Node load balancing deployment method for spatial network simulation

Abstract

The invention discloses a node load balancing deployment method for spatial network simulation, and belongs to the technical field of network simulation. Firstly, a spatial network simulation node deployment model is established, and input, output and resource constraint models of a node load balancing deployment method are defined; then, an objective function of a node load balancing deployment method is constructed through the deployment model, a resource matching degree concept is introduced to balance simulation node loads, and a solution space of the deployment method is generated through a pheromone mechanism and probability scheduling; secondly, for a solution strategy in the solution space, verifying the satisfaction condition of the solution strategy to the resource constraint model, and comparing and updating the optimal solution and the optimal scheme according to the satisfaction condition; and finally, the updated solution strategy is utilized to refresh pheromones of the deployment method, iteration is carried out until a termination condition is reached, and a final optimal objective function value and an optimal deployment scheme are output. According to the invention, the placement of the spatial network simulation nodes can be balanced, and extra resource overhead can be reduced as much as possible under the condition of ensuring the simulation service.

Description

Technical Field

[0001] The present invention relates to the technical field of network simulation, and in particular to a node load balancing deployment method for space network simulation. Background Art

[0002] In recent years, large-scale satellite constellation deployment plans have been proposed from time to time, such as the OneWeb constellation and the Starlink constellation; combined with ground-based nodes, the total number of nodes in the entire scenario can reach thousands or even tens of thousands. The simulation of large-scale scenarios is an unavoidable problem in space network simulation. However, the currently more commonly used single-machine network simulation method based on virtualization technology is limited by computer resources. When facing target scenarios with many nodes, it may encounter problems with insufficient hardware resources and cannot complete the simulation. Therefore, in space network simulation, hardware resources are generally expanded by using multiple servers or clusters, and hardware resources are generally managed and allocated through a container orchestration system.

[0003] Traditional node load balancing deployment methods generally only consider the relationship between the hardware resources required by the simulation node and the remaining hardware resources of the server where the simulation node is deployed to place the node. The widely used container orchestration system kubernetes (k8s) adopts load balancing placement strategies such as MostRequestPriority and BalancedResouceAllocation in the node placement stage. The key factors considered are the relationship between the CPU, memory, disk and other resources required by the simulation node container to be placed and the remaining CPU, memory and disk resources available on the server. This is suitable for cloud node placement scenarios where the goal is to provide services for requests outside the cluster. However, for the space network simulation scenario, its main business is to simulate the space network function. In addition to the fact that each simulation node requires hardware resources such as CPU, memory, disk to meet the normal operation of its internal processes and services, the simulation space network function also includes another very important aspect, namely the simulation of space network links. In space networks, since satellites are constantly moving, the connection relationship between satellite nodes, link characteristics, network topology, etc. all change over time; however, each satellite node operates periodically according to a predetermined orbit, and the visibility relationship between satellite nodes is also determined over time. Therefore, according to the connection relationship between satellite nodes, the network topology is also determined in each time slot. In order to simulate space network links that change over time, it is necessary to establish link connections between simulation nodes in each time slot.

[0004] In space network simulation, a multi-layer virtual switch (OVS) is usually used to establish links and forward data between simulation nodes. OVS needs to rely on the CPU to process the exchange and forwarding of data packets. Therefore, the resources occupied by OVS are ultimately manifested as additional occupation of server CPU resources. Therefore, in space network simulation, in addition to the CPU, memory, disk and other resources required by the simulation nodes to meet the normal operation of their internal processes and services, a large amount of CPU resources are additionally occupied due to the network data interaction on the space network links flexibly established between simulation nodes. In addition, since there will be a large number of simulation links between simulation nodes in space network simulation, and these simulation nodes are likely to be deployed to different bearer servers, a large amount of data interaction between different servers will be brought about at this time, occupying the bandwidth resources of the server. Therefore, placing space network simulation nodes according to the original cloud node placement load balancing method, only considering the relationship between the demand and remaining resources of the simulation nodes and the server, while ignoring the consideration of the space network simulation links, will inevitably cause a series of unreasonable deployment problems, and may even affect the preset bandwidth of the simulation link due to improper node placement, resulting in insufficient remaining resources on the server to support the continuation of the simulation business, thereby affecting the space network simulation effect.

[0005] Therefore, in order to achieve accurate spatial network simulation in each time slot, in the node load balancing deployment method, it is far from enough to only consider the hardware resources required by the simulation nodes and the hardware resources that the server can provide. The additional CPU resources occupied by the establishment of simulation links in the spatial network simulation and the communication bandwidth between servers should also be considered. Summary of the invention

[0006] In view of the problem that the current node load balancing deployment method applied in space network simulation lacks consideration of the large amount of additional resources occupied by network data interaction on the space network simulation link, the present invention provides a node load balancing deployment method for space network simulation, which aims to take into account the additional resource requirements caused by the space network simulation link, balance the placement of space network simulation nodes, and reduce the additional overhead of all servers as much as possible while ensuring the simulation business.

[0007] The above purpose is achieved through the following technical solutions: The present invention provides a node load balancing deployment method for space network simulation, comprising the following steps: S1. Establish a space network simulation node deployment model: Analyze the particularity of the space network simulation scenario, build the basic framework of the space network simulation node deployment model, clarify the input and output content of the node load balancing deployment method, and clarify the hardware resource constraint model.

[0008] S2. According to the space network simulation node deployment model established in step S1, the objective function of the space network simulation node load balancing deployment method is constructed, the concept of resource matching is introduced to balance the simulation node load, and the solution space of the space network simulation node deployment method is generated by using the pheromone mechanism and the idea of ​​probability scheduling.

[0009] S3. For the solution in the solution space generated in step S2, verify whether it satisfies the hardware resource constraint model, and compare and update the solution in the solution space of step S2 with the recorded optimal solution and optimal plan based on the result.

[0010] S4. Use the solution strategy updated in step S3 to refresh the pheromone of the spatial network simulation node deployment model, continue to iterate and generate the solution space until the termination condition is reached, and output the final optimal objective function value and the optimal deployment plan.

[0011] Furthermore, the step S1 specifically includes the following sub-steps: S1.1 Analyze the particularity of space network simulation scenarios based on the characteristics of space network simulation. During space network simulation, in addition to the necessary hardware resource requirements to maintain the normal operation of its internal services, simulation nodes also need to simulate space network links.

[0012] S1.2 Construct the basic framework of the space network simulation node deployment model and provide the input and output content of the node load balancing deployment method.

[0013] The space network simulation node deployment model includes A collection of simulation nodes And contains Server collection of servers .

[0014] ; ; in, Represents the simulation nodes , Represents the server .

[0015] For the simulation links between simulation nodes in space network simulation based on OVS, a simulation node connection relationship matrix is ​​used It means to express the connection relationship between simulation nodes: ; matrix middle Represented as a simulation node Simulation Node If the simulation node Simulation Node If there is no link connection between 0, simulation node Simulation Node If there is a link connection between is 1.

[0016] Defining variables Represents a simulation node Whether to deploy on the server If the simulation node Deploy on the server ,but ; Otherwise it is 0. Since a simulation node can only be deployed on one server, .

[0017] The input of the deployment model is a collection of simulation nodes , Server Collection And the simulation node connection relationship matrix The output of the deployment model is the minimum value of the objective function value of the deployment plan, recorded as the optimal objective function value, and the simulation node deployment plan when the objective function value takes the minimum value, recorded as the optimal deployment plan.

[0018] S1.3 Define the resource constraint model. The hardware resources that can be provided are represented as a three-dimensional vector : ; in, , and The server Able to provide CPU, memory, and disk resources for simulation nodes.

[0019] Because the server is still running some processes before deploying nodes, which consumes some resources, it is necessary to quantify the resources already used by the server. The resource load is defined as a three-dimensional vector : ; in, , and Is the server Used CPU, memory, and disk resources.

[0020] For each simulation node In order to maintain the normal operation of its internal business, the demand for server resources is defined as the following three-dimensional vector : ; in, , and Represents the simulation nodes For CPU, memory, and disk resources.

[0021] In order to achieve the purpose of load balancing, the sum of all hardware requirements of the simulation nodes loaded by each server should not exceed the available hardware resources of the server. Therefore, the final resource constraint model is expressed as the sum of the resources used by the server and the resources used for the simulation nodes cannot exceed the resources it can provide. The resource constraint model is as follows: ;

[0022] in Indicates the preset bandwidth of the simulated link; The additional CPU resource requirements on the server to meet the preset bandwidth of the emulated link; Representation Server The available bandwidth of the network interface.

[0023] Step S2 specifically includes the following sub-steps:

[0024] S2.1 According to the spatial network simulation node deployment model in step S1, construct the objective function of the node load balancing deployment method of spatial network simulation. In order to use as few hardware resources as possible to complete the placement of simulation nodes and the construction of simulation scenarios in spatial network simulation, the final objective function can be obtained as follows: Under the premise of not exceeding the available CPU, memory, disk and network port communication bandwidth of each server, the additional CPU resource usage caused by the simulation link between simulation nodes relying on OVS is minimized. , expressed as: .

[0025] S2.2 introduces the concept of resource matching to balance the load of simulation nodes.

[0026] The concept of resource matching is divided into three parts: CPU resource matching, memory resource matching, and disk resource matching.

[0027] CPU resource matching It is expressed as the ratio of the CPU resources that the server can provide to the CPU resource requirements of the simulation node.

[0028] ; in, Is the server CPU resources used; Is the server The CPU resources available; Represents a simulation node The demand for CPU resources.

[0029] Memory resource matching It is expressed as the ratio of the memory resources that the server can provide to the memory resource requirements of the simulation node.

[0030] ; in, Is the server Memory resources used; Is the server The memory resources available; Represents a simulation node The demand for memory resources.

[0031] Disk resource matching It is expressed as the ratio of the disk resources that the server can provide to the disk resources required by the simulation node.

[0032] ; in, Is the server Disk resources used; Is the server The disk resources that can be provided; Represents a simulation node The demand for disk resources.

[0033] The CPU resource matching degree, memory resource matching degree, and disk resource matching degree are weighted according to their importance. If the final resource matching degree is greater, it proves that the potential benefit of deploying the simulation node to the server is greater, and it is more likely to be deployed to the server. Therefore, the final resource matching degree is It is expressed as the weighted average of these three.

[0034] S2.3 Generate the solution space of the deployment method using the pheromone mechanism and the idea of ​​probabilistic scheduling. The addition of pheromones is based on the solution found by the current search individual, and the pheromone concentration on the relevant path is updated according to the quality of the solution. The pheromone concentration on the path with better solution increases faster, thereby guiding more individuals to choose these paths in the next iteration. The solution construction process is a probabilistic process. Each individual in the search population chooses the next path based on the combined influence of the pheromone concentration on the path and the heuristic factor. In the process of solution construction, the simulation node According to probability Dispatched to the server : ; in and They represent the search for individual deployment simulation nodes at the iteration number 𝑡. To the server The matching degree of pheromone and resources at that time; and are the weight parameters of pheromone and resource matching degree respectively. The higher the pheromone, the more simulation nodes are deployed. To the server The higher the resource matching degree, the stronger the tendency of the simulation node. Assign to server The higher the potential benefit, the stronger the preference.

[0035] For each individual in the search population, a solution strategy will be generated based on the pheromone mechanism and probability scheduling. The solution strategy includes the objective function value and the corresponding simulation node deployment plan. The collection of all solution strategies constitutes the solution space of this search.

[0036] Step S3 specifically includes the following sub-steps: S3.1 Constraint verification: For the solution space obtained in S2, it is necessary to verify whether each solution strategy satisfies the constraints of the resource constraint model in step S1.

[0037] If the simulation node deployment plan of the solution strategy does not meet the restrictions of the resource constraint model in step S1, the objective function value of the solution strategy is modified to a penalty value, and the simulation node deployment plan of the solution strategy is abandoned to avoid the final optimal deployment plan not meeting the constraints.

[0038] S3.2 Compare and update the optimal solution and the optimal plan. If the solution strategy constructed by the verification meets the constraint conditions, the objective function value obtained by this solution strategy is compared with the optimal solution at this time. If its value is less than the optimal solution, replace the objective function value obtained by this solution strategy with the optimal solution, and record the simulation node deployment plan at this time as the optimal plan: ; in Represented as a simulation node Deploy on the server And so on. This is the simulation node deployment plan under this solution strategy.

[0039] Step S4 is the following steps: S4.1 Refresh the pheromone of the deployment model using the updated solution strategy. The pheromone update process is divided into two steps: pheromone volatilization and pheromone addition. The mathematical expression of pheromone volatilization is as follows: ; in, Represents iteration Secondary simulation node Deploy to the server The pheromone concentration on. is the pheromone volatilization factor, which usually takes a value between 0 and 1. Through this operation, the pheromone concentration will be multiplied by a factor less than 1 after each iteration, so it will gradually decrease.

[0040] The addition of pheromones is based on the solution found by the current search individual, and the pheromone concentration on the relevant path is updated according to the quality of the solution. The pheromone concentration on the path with better solutions increases faster, thereby guiding more search individuals to choose these paths. Each search individual adds pheromones to the path it passes through, and the mathematical expression is as follows: ; in: is the pheromone update constant, usually recorded as 1, Represents a search for an individual The cost of the solution found, numInd Represents the total number of search individuals in the search population. For better solutions, the lower the cost, the greater the increment of pheromone. Therefore, the objective function value of the solution strategy constructed using this search individual is taken as More appropriately, a lower objective function value means that the solution strategy is optimal.

[0041] Combining pheromone volatilization and pheromone addition, the complete pheromone update is expressed as: .

[0042] S4.2 Continue to iterate and generate the solution space until the termination condition is reached, and output the final optimal objective function value and the optimal deployment plan. After step S4.1 is completed, according to the pheromone of the refreshed deployment method, steps S2, S3 and S4.1 are repeated until the preset iteration limit is reached. After reaching it, the final optimal solution is output as the optimal objective function value , and the corresponding optimal solution is the optimal deployment solution : ; in Represents the simulation node in the optimal deployment solution Deploy on the server And so on.

[0043] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a node load balancing deployment method for space network simulation. The method can more reasonably deploy space network simulation nodes to appropriate servers, solving the problem that the remaining server resources that may occur in space network simulation are insufficient to support simulation services, affecting the space network simulation effect, etc. Under certain server resources, when a certain amount of simulation nodes are deployed, the additional overhead of all servers is reduced as much as possible while ensuring the simulation services, thereby improving the efficiency and reliability of the simulation services, which has certain promotion significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of the node load balancing deployment method for space network simulation Figure 2 It is a typical structure of space network simulation Figure 3 This is a diagram showing the connection relationship of the Iridium simulation scenario according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The embodiments of the present invention are further described below in conjunction with the accompanying drawings and specific implementation methods.

[0046] Reference Figure 1 , introduces the node load balancing deployment method process for space network simulation. The whole process can be divided into the following four steps: S1. Establish a space network simulation node deployment model: Analyze the particularity of the space network simulation scenario, build the basic framework of the space network simulation node deployment model, clarify the input and output content of the node load balancing deployment method, and clarify the hardware resource constraint model.

[0047] S2. According to the space network simulation node deployment model established in step S1, the objective function of the space network simulation node load balancing deployment method is constructed, the concept of resource matching is introduced to balance the simulation node load, and the solution space of the space network simulation node deployment method is generated by using the pheromone mechanism and the idea of ​​probability scheduling.

[0048] S3. For the solution in the solution space generated in step S2, verify whether it satisfies the hardware resource constraint model, and compare and update the solution in the solution space of step S2 with the recorded optimal solution and optimal plan based on the result.

[0049] S4. Use the solution strategy updated in step S3 to refresh the pheromone of the spatial network simulation node deployment model, continue to iterate and generate the solution space until the termination condition is reached, and output the final optimal objective function value and the optimal deployment plan.

[0050] Furthermore, the step S1 specifically includes the following sub-steps: S1.1 Analyze the particularity of space network simulation scenarios based on the characteristics of space network simulation. During space network simulation, in addition to the necessary hardware resource requirements to maintain the normal operation of its internal business, the simulation node also needs to simulate the space network link. The simulation space network link relies on the layer virtual switch (OVS). OVS needs to rely on the CPU to process the switching and forwarding of data packets, so the resources occupied by OVS ultimately manifest as additional occupation of server CPU resources. Therefore, in space network simulation, in addition to the CPU, memory, disk and other resources required by the simulation node to meet the normal operation of its internal processes and businesses, a large amount of CPU resources are additionally occupied due to the network data interaction on the space network link flexibly established between simulation nodes.

[0051] See attached Figure 2 The typical structure of space network simulation. Different node deployment strategies will lead to different additional hardware resource overhead caused by simulating space network links. Specifically, there are two cases: If the two simulation nodes at both ends of a spatial network simulation link are deployed on the same server (such as the attached Figure 2 The simulation link formed by these two simulation nodes only requires CPU resources for one OVS forwarding, and does not involve communication bandwidth between servers. Therefore, the preset bandwidth of the simulation link between simulation nodes is set to , then in order to meet this preset bandwidth, OVS needs to occupy Therefore, if the two simulation nodes at both ends of a simulation link are deployed on the same server, the additional hardware resource requirements for the server are CPU resources.

[0052] If the two simulation nodes at both ends of a spatial network simulation link are deployed on different servers (such as the attached Figure 2 In the simulation node 1 and simulation node 4, the OVS in each server needs to forward the communication traffic of the simulation link, and the physical link connecting the two servers also needs to forward the traffic. Therefore, in order to meet the preset bandwidth of this simulation link , not only does it take up space on each server CPU core resources, and also occupy the communication bandwidth between servers Therefore, if the two simulation nodes at both ends of a simulation link are deployed on different servers, the additional hardware resource requirements for each server where the two simulation nodes are located are CPU resources and bandwidth resources.

[0053] Therefore, in addition to the resources required to maintain its internal business, the simulation nodes in the space network simulation will also need more additional resources to meet the needs of its space network simulation links, which is its special feature.

[0054] Reference Figure 3 Iridium network simulation scenario connection relationship diagram, the space network simulation scenario of this embodiment is an Iridium communication scenario, which includes 66 simulation nodes, belonging to 6 satellite orbits. As shown in the figure, the 6 orbits are separated by dotted lines. On each orbit, 11 Iridium satellites are running, which are simulation nodes in this embodiment. On each orbit, adjacent satellites are connected end to end. Satellites adjacent to each other on the left and right on different orbits are connected to each other on the left and right. It should be noted that due to the existence of the reverse seam, the leftmost orbit and the rightmost orbit in the figure, that is, the satellites adjacent to each other on the left and right on the 1st orbit and the 6th orbit do not establish a link connection.

[0055] S1.2 Construct the basic framework of the space network simulation node deployment model and provide the input and output content of the node load balancing deployment method.

[0056] The space network simulation node deployment model includes A collection of simulation nodes And contains Server collection of servers .

[0057] ; ; in, Represents the simulation nodes , Represents the server .

[0058] In the Iridium network simulation scenario of this embodiment, the number of simulation nodes is 66 and the number of bearer servers is 3.

[0059] For the simulation links between simulation nodes in space network simulation based on OVS, a simulation node connection relationship matrix is ​​used It means to express the connection relationship between simulation nodes: ; matrix middle Represented as a simulation node Simulation Node If the simulation node Simulation Node If there is no link connection between 0, simulation node Simulation Node If there is a link connection between is 1.

[0060] Defining variables Represents a simulation node Whether to deploy on the server If the simulation node Deploy on the server ,but ; Otherwise it is 0. Since a simulation node can only be deployed on one server, .

[0061] The input of the deployment model is a collection of simulation nodes , Server Collection And the simulation node connection relationship matrix The output of the deployment model is the minimum value of the objective function value of the deployment plan, recorded as the optimal objective function value, and the simulation node deployment plan when the objective function value takes the minimum value, recorded as the optimal deployment plan.

[0062] S1.3 Clarify the resource constraint model. To meet the needs of server load balancing, each server is specified The hardware resources that can be provided are represented as a three-dimensional vector : ; in, , and The server Able to provide CPU, memory, and disk resources for simulation nodes.

[0063] Because the server is still running some processes before deploying nodes, which consumes some resources, it is necessary to quantify the resources already used by the server. The resource load is defined as a three-dimensional vector : ; in, , and Is the server Used CPU, memory, and disk resources.

[0064] For each simulation node In order to maintain the normal operation of its internal business, the demand for server resources is defined as the following three-dimensional vector : ; in, , and Represents the simulation nodes For CPU, memory, and disk resources.

[0065] In this embodiment, each server CPU has 36 cores, 64G memory and 512G disk resources are installed. Without running other processes, these resources can be almost fully provided for space network simulation. Each simulation node has a fixed requirement of 1 core, 1.5G memory and 3G disk resources. After testing, in this embodiment, the preset bandwidth between simulation nodes is 100 To reserve bandwidth, the server needs to occupy 0.1 of the CPU core resources. Each server has a 10G network card, which can provide 10,000 bandwidth.

[0066] In order to achieve the purpose of load balancing, the sum of all hardware requirements of the simulation nodes loaded by each server should not exceed the available hardware resources of the server. Therefore, the final resource constraint model is expressed as the sum of the resources used by the server and the resources used for the simulation nodes cannot exceed the resources it can provide. The resource constraint model is as follows: ; in, is the number of simulation nodes; The number of servers that host simulation tasks; Indicates the preset bandwidth of the simulated link; The additional CPU resource requirements on the server to meet the preset bandwidth of the emulated link; Representation Server The available bandwidth of the network interface.

[0067] Step S2 specifically includes the following sub-steps:

[0068] S2.1 According to the spatial network simulation node deployment model in step S1, construct the objective function of the node load balancing deployment method of spatial network simulation. In order to use as few hardware resources as possible to complete the placement of simulation nodes and the construction of simulation scenarios in spatial network simulation, the ultimate goal can be obtained: that is, to minimize the additional CPU resource usage caused by the simulation links between simulation nodes relying on OVS without exceeding the available CPU, memory, disk and network port communication bandwidth of each server. , expressed as: .

[0069] S2.2 Introducing the concept of resource matching to balance the load of simulation nodes. In the process of placing simulation nodes, for nodes with more resource consumption, placing them in servers with more remaining resources can avoid excessive resource usage of a certain server and achieve the purpose of load balancing. Therefore, the concept of resource matching is introduced to balance the load of simulation nodes.

[0070] The concept of resource matching is divided into three parts: CPU resource matching, memory resource matching, and disk resource matching.

[0071] CPU resource matching It is expressed as the ratio of the CPU resources that the server can provide to the CPU resource requirements of the simulation node: ; in, Is the server CPU resources used; Is the server The CPU resources available; Represents a simulation node The demand for CPU resources.

[0072] Memory resource matching It is expressed as the ratio of the memory resources that the server can provide to the memory resource requirements of the simulation node: ; in, Is the server Memory resources used; Is the server The memory resources available; Represents a simulation node The demand for memory resources.

[0073] Disk resource matching It is expressed as the ratio of the disk resources that the server can provide to the disk resources required by the simulation node: ; in, Is the server Disk resources used; Is the server The disk resources that can be provided; Represents a simulation node The demand for disk resources.

[0074] The three are weighted according to a certain degree of importance. If the final resource matching degree is greater, it proves that the potential benefit of deploying the simulation node to the server is greater, and it is more likely to be deployed to the server. Such a deployment scheme can avoid the simulation node with large resource requirements from being dispatched to the server with limited remaining resources, balancing the server load. Ultimately, a better load balancing effect is achieved. Therefore, the final resource matching degree is It is expressed as the weighted average of the three. In the spatial network simulation, we pay more attention to the guarantee of CPU resources, so its weight is set to the highest, followed by memory and hard disk resources.

[0075] In this embodiment, more emphasis is placed on the guarantee of the CPU resources of the simulation node by the server, so its weight is set to the highest, followed by memory and hard disk resources; .

[0076] S2.3 Generate the solution space of the deployment method using the pheromone mechanism and the idea of ​​probabilistic scheduling. The addition of pheromones is based on the solution found by the current search individual, and the pheromone concentration on the relevant path is updated according to the quality of the solution. The pheromone concentration on the path with better solution increases faster, thereby guiding more individuals to choose these paths in the next iteration. The solution construction process is a probabilistic process. Each individual in the search population chooses the next path based on the combined influence of the pheromone concentration on the path and the heuristic factor. In the process of solution construction, the simulation node According to probability Dispatched to the server : ; in and They represent the search for individual deployment simulation nodes at the iteration number 𝑡. To the server The matching degree of pheromone and resources at that time; and are the weight parameters of pheromone and resource matching degree respectively. The higher the pheromone, the more simulation nodes are deployed. To the server The higher the resource matching degree, the stronger the tendency of the simulation node. Assign to server The higher the potential benefit, the stronger the preference.

[0077] In this embodiment, the weight parameter of the pheromone Set to 1, because more emphasis is placed on load balancing of simulation nodes, the weight parameter of resource matching is Set it to a higher value of 5 to ensure that all servers can evenly distribute all simulation nodes.

[0078] For each individual in the search population, a solution strategy will be generated based on the pheromone mechanism and probability scheduling. The solution strategy includes the objective function value and the corresponding simulation node deployment plan. The collection of all solution strategies constitutes the solution space of this search.

[0079] Step S3 specifically includes the following sub-steps: S3.1 Constraint verification: For the solution space obtained in S2, it is necessary to verify whether each solution strategy satisfies the constraints of the resource constraint model in step S1.

[0080] If the simulation node deployment plan of the solution strategy does not meet the restrictions of the resource constraint model in step S1, the objective function value of the solution strategy is modified to a large penalty value, and the simulation node deployment plan of the solution strategy is abandoned to avoid the final optimal deployment plan not meeting the constraints.

[0081] In this embodiment, the penalty value when the simulation node deployment scheme of the solution strategy does not meet the restrictions of the resource constraint model in step S1 is set to 100000 to ensure that the simulation node deployment scheme of the solution strategy is abandoned.

[0082] S3.2 Compare and update the optimal solution and the optimal plan. If the solution strategy constructed by the verification meets the constraint conditions, the objective function value obtained by this solution strategy is compared with the optimal solution at this time. If its value is less than the optimal solution, replace the objective function value obtained by this solution strategy with the optimal solution, and record the simulation node deployment plan at this time as the optimal plan: ; in Represented as a simulation node Deploy on the server And so on. This is the simulation node deployment plan under this solution strategy.

[0083] Step S4 is the following steps: S4.1 Refresh the pheromone of the deployment method using the updated solution strategy. The pheromone update process can be divided into two steps: pheromone volatilization and pheromone addition. The volatilization process ensures that the pheromone concentration of the old, suboptimal path gradually decreases and does not affect subsequent searches. The mathematical expression of pheromone volatilization is as follows: ; in: Represents iteration Secondary simulation node Deploy to the server The pheromone concentration on. is the pheromone volatilization factor, which usually takes a value between 0 and 1. Through this operation, the pheromone concentration will be multiplied by a factor less than 1 after each iteration, so it will gradually decrease.

[0084] In this embodiment, the pheromone volatile factor Set it to 0.1 to avoid sudden changes in solution strategy.

[0085] The addition of pheromones is based on the solution found by the current search individual, and the pheromone concentration on the relevant path is updated according to the quality of the solution. The pheromone concentration on the path with better solutions increases faster, thereby guiding more search individuals to choose these paths. Each search individual adds pheromones to the path it passes through, and the mathematical expression is as follows: ; in: is the pheromone update constant, usually recorded as 1, Represents a search for an individual The cost of the solution found, numInd Represents the total number of search individuals in the search population. For better solutions, the lower the cost, the greater the increment of pheromone. Therefore, the objective function value of the solution strategy constructed using this search individual is taken as More appropriately, a lower objective function value means that the solution strategy is optimal.

[0086] Combining the above two parts, the complete pheromone update is expressed as: .

[0087] S4.2 Continue to iterate and generate the solution space until the termination condition is reached, and output the final optimal objective function value and the optimal deployment plan. After step S4.1 is completed, according to the pheromone of the refreshed deployment method, steps S2, S3 and S4.1 are repeated until the preset iteration limit is reached. After reaching it, the final optimal solution is output as the optimal objective function value , and the corresponding optimal solution is the optimal deployment solution : ; in Represents the simulation node in the optimal deployment solution Deploy on the server And so on.

[0088] Finally, the result data of this embodiment is shown in Table 1: Table 1 Results of this example

[0089] It can be seen from Table 1 that, in this embodiment, compared with the traditional node average distribution deployment method, the deployment method of this solution can save the additional CPU resource occupancy caused by the simulation link traffic forwarding through OVS, reducing the additional occupancy by about 16.7%. At the same time, due to the more reasonable simulation node placement, the network port communication pressure between servers is also significantly reduced, and the demand for total simulation bandwidth has shown a significant decrease.

Claims

1. A node load balancing deployment method for space network simulation, characterized in that the steps include: S1. Establish a space network simulation node deployment model: Build the basic framework of the space network simulation node deployment model, clarify the input and output content of the node load balancing deployment method, and clarify the hardware resource constraint model; S2. According to the space network simulation node deployment model established in step S1, the objective function of the space network simulation node load balancing deployment method is constructed, the concept of resource matching is introduced to balance the simulation node load, and the solution space of the space network simulation node deployment method is generated by using the pheromone mechanism and the idea of ​​probabilistic scheduling; S3. For the solution in the solution space generated in step S2, verify whether it satisfies the hardware resource constraint model, and compare and update the solution in the solution space of step S2 with the recorded optimal solution and optimal solution according to the result; S4. Refresh the pheromone of the spatial network simulation node deployment model using the solution strategy updated in step S3, continue to iterate and generate the solution space until the termination condition is reached, and output the final optimal objective function value and the optimal deployment plan.

2. A node load balancing deployment method for space network simulation according to claim 1, characterized in that The specific implementation steps of step S1 include: S1.1 According to the characteristics of space network simulation, the particularity of space network simulation scenarios is analyzed; in the process of space network simulation, in addition to the necessary hardware resource requirements to maintain the normal operation of its internal business, the simulation node also needs to simulate the space network link; S1.2 Construct the basic framework of the space network simulation node deployment model and provide the input and output content of the node load balancing deployment method; The space network simulation node deployment model includes A collection of simulation nodes And contains Server collection of servers ; ； ； in, Represents the simulation nodes , Represents the server ; For the simulation links between simulation nodes in space network simulation based on OVS, a simulation node connection relationship matrix is ​​used It means to express the connection relationship between simulation nodes: ； matrix middle Represented as a simulation node Simulation Node If the link connection relationship between the simulation nodes Simulation Node If there is no link connection between 0, simulation node Simulation Node If there is a link connection between is 1; Defining variables Represents a simulation node Whether to deploy on the server If the simulation node Deploy on the server ,but ; Otherwise it is 0. Since a simulation node can only be deployed on one server, ; The input of the deployment model is a collection of simulation nodes , Server Collection And the simulation node connection relationship matrix , the output of the deployment model is the minimum value of the objective function value of the deployment plan, recorded as the optimal objective function value, and the simulation node deployment plan when the objective function value takes the minimum value, recorded as the optimal deployment plan; S1.3 Define the resource constraint model; specify the hardware resources that each server s can provide, represented as a three-dimensional vector : ； in, , and The server The CPU, memory, and disk resources that can be provided to the simulation nodes; Because the server is still running some processes before deploying nodes, which consumes some resources, it is necessary to quantify the resources already used by the server. The resource load is defined as a three-dimensional vector : ； in, , and The server CPU, memory, and disk resources used; For each simulation node In order to maintain the normal operation of its internal business, the demand for server resources is defined as the following three-dimensional vector , expressed as the following formula: ； in, , and Represents the simulation nodes For CPU, memory, and disk resources; In order to achieve the purpose of load balancing, the sum of all hardware requirements of the simulation nodes loaded by each server should not exceed the available hardware resources of the server. Therefore, the final resource constraint model is expressed as the sum of the resources used by the server and the resources used for the simulation nodes cannot exceed the resources it can provide. The resource constraint model is as follows: ； in, is the number of simulation nodes; The number of servers that host simulation tasks; Indicates the preset bandwidth of the simulated link; The additional CPU resource requirements on the server to meet the preset bandwidth of the emulated link; Representation Server The available bandwidth of the network interface.

3. A node load balancing deployment method for space network simulation according to claim 2, characterized in that The specific implementation steps in step S2 include: S2.1 According to the spatial network simulation node deployment model in step S1, the objective function of the node load balancing deployment method of spatial network simulation is constructed. The objective function is to minimize the additional CPU resource usage caused by the simulation link between simulation nodes relying on OVS without exceeding the available CPU, memory, disk and network port communication bandwidth of each server. , expressed as: ； S2.2 Introducing the concept of resource matching to balance the load of simulation nodes. The concept of resource matching is divided into three parts: CPU resource matching, memory resource matching, and disk resource matching; CPU resource matching It is expressed as the ratio of the CPU resources that the server can provide to the CPU resource requirements of the simulation node: ； in, Is the server CPU resources used; Is the server The CPU resources available; Represents a simulation node The demand for CPU resources; Memory resource matching It is expressed as the ratio of the memory resources that the server can provide to the memory resource requirements of the simulation node: ； in, Is the server Memory resources used; Is the server The memory resources available; Represents a simulation node The need for memory resources; Disk resource matching It is expressed as the ratio of the disk resources that the server can provide to the disk resources required by the simulation node: ； in, Is the server Disk resources used; Is the server The disk resources that can be provided; Represents a simulation node The demand for disk resources; The CPU resource matching degree, memory resource matching degree, and disk resource matching degree are weighted according to their importance. If the final resource matching degree is greater, it proves that the potential benefit of deploying the simulation node to the server is greater, and it is more likely to be deployed to the server. Therefore, the final resource matching degree is It is expressed as the weighted average of these three; S2.3 Generate the solution space of the deployment method using the pheromone mechanism and the idea of ​​probabilistic scheduling. The addition of pheromones is based on the solution found by the current search individual. The pheromone concentration on the relevant path is updated according to the quality of the solution. The pheromone concentration on the path with better solution increases faster, thereby guiding more individuals to choose these paths in the next iteration. The solution construction process is a probabilistic process. Each individual in the search population chooses the next path based on the combined influence of the pheromone concentration on the path and the heuristic factor. In the process of solution construction, the simulation node According to probability Dispatched to the server : ； in, and They represent the search for individual deployment simulation nodes at the iteration number 𝑡. To the server The matching degree of pheromone and resources at that time; and are the weight parameters of pheromone and resource matching degree respectively. The higher the pheromone, the more simulation nodes are deployed. To the server The stronger the tendency, the higher the resource matching degree, and the simulation node Assign to server The higher the potential benefit, the stronger the tendency; For each individual in the search population, a solution strategy will be generated based on the pheromone mechanism and probability scheduling. The solution strategy includes the objective function value and the corresponding simulation node deployment plan. The collection of all solution strategies constitutes the solution space of this search.

4. A node load balancing deployment method for space network simulation according to claim 3, characterized in that The specific implementation steps of step S3 include: S3.1 Constraint verification: For the solution space obtained in S2, it is necessary to verify whether each solution strategy satisfies the constraints of the resource constraint model in step S1; If the simulation node deployment plan of the solution strategy does not meet the restrictions of the resource constraint model in step S1, the objective function value of the solution strategy is modified to a penalty value, and the simulation node deployment plan of the solution strategy is abandoned to avoid the final optimal deployment plan not meeting the constraint conditions; S3.2 The optimal solution and the optimal plan are compared and updated; if the constructed solution strategy is verified to meet the constraints, the objective function value obtained by this solution strategy is compared with the optimal solution at this time. If its value is less than the optimal solution, the objective function value obtained by this solution strategy is replaced as the optimal solution, and the simulation node deployment plan at this time is recorded as the optimal plan: ； in, Represented as a simulation node Deploy on the server On, and so on, This is the simulation node deployment plan under this solution strategy.

5. A node load balancing deployment method for space network simulation according to claim 1, characterized in that The specific implementation steps of step S3 include: S4.1 Use the updated solution strategy to refresh the pheromone of the deployment model. The pheromone update process is divided into two steps: pheromone volatilization and pheromone addition. The mathematical expression of pheromone volatilization is as follows: ； in, Represents iteration Secondary simulation node Deploy to the server The pheromone concentration on Represents iteration Secondary simulation node Deploy to the server The pheromone concentration on is the pheromone volatilization factor, which usually takes a value between 0 and 1. Through this operation, the pheromone concentration will be multiplied by a factor less than 1 after each iteration, so it will gradually decrease; The addition of pheromones is based on the solution found by the current search individual. The pheromone concentration on the relevant path is updated according to the quality of the solution. The pheromone concentration on the path with better solution increases faster, thereby guiding more search individuals to choose these paths. Each search individual adds pheromones to the path it passes through. The mathematical expression is as follows: ； in: is the pheromone update constant, usually recorded as 1, Represents a search for an individual The cost of the solution found, numInd represents the total number of search individuals in the search population; Combining pheromone volatilization and pheromone addition, the complete pheromone update is expressed as: ； S4.2 Continue to iterate and generate the solution space until the termination condition is reached, and output the final optimal objective function value and the optimal deployment plan. After step S4.1 is completed, according to the pheromone of the refreshed deployment method, steps S2, S3 and S4.1 are repeated until the preset iteration limit is reached. After reaching it, the final optimal solution is output as the optimal objective function value. , and the corresponding optimal solution is the optimal deployment solution : ； in Represents the simulation node in the optimal deployment solution Deploy on the server And so on.

Images