Traditional methods mostly realize load shunting of communication system through specific hardware, but the scalability and sensitivity of hardware products restrict the quality of load shunting, and the system is in danger of load imbalance. Therefore, in order to improve the load balance of electronic communication system, a shunt controller of electronic communication system is designed and implemented based on dynamic Hash algorithm. The shunt controller uses Telnet component to deploy the router remotely, uses SNMP component to obtain the traffic in the router, manages the current variables, and uses TFTP server to feedback command files to the router to ensure the traffic of the router interface. Equalization. The shunt controller completes the shunt control of the electronic communication system by dynamic updating and balancing module, preprocessing module and balancing shunt module, and adopts the dynamic Hash algorithm based on the flow to achieve the system flow balancing. The key code of shunt controller shaping policy routing is given. The experimental results show that the three indexes of transmission delay, transmission rate and shunt error of the electronic communication system under the shunt controller designed are better, and the optimization of network load balancing is realized. With the rapid development of communication technology, the application of electronic communication system in people’s production and life is gradually expanding. Because of the rapid growth of the bandwidth of electronic communication network, in order to enhance the efficiency of IP message acquisition, analysis and operation of communication system routers, it is necessary to use multi-processors to coordinate message operation, which plays an important role in load balancing of multi-processors. Traditional methods mostly realize load shunting of communication system through specific hardware, but the scalability and sensitivity of hardware products restrict the quality of load shunting of system [2]. Through the shunt controller software, the IP message shunt of the router in the electronic communication system under its shunt state is realized, and the load balance of the system is ensured [3?4]. There are many drawbacks in the current classification methods. For example, in document [5], the rotation method is used to complete the load control of communication system traffic, and the application is sent to different servers in turn by the rotation method. But this method is relatively simple, and there is a long time to apply for services. Literature [6] Schedules network traffic through the minimum connection method, which analyses the server’s load status based on the number of connections currently working on the server. However, it consumes too much system resources to effectively describe the actual system application load. Document [7] Through the balancing component of the minimum missing method, the application status of the storage node, the next application is transmitted to the node with the lowest application for historical operation to realize the load balancing of the system. This method needs to detect the node through the special technology of UDP package, which has great limitations. In reference [8], the system resource load method based on weighted method and rotation method is proposed. The load balancing queue is modeled according to the priority and load state of nodes. The load balancing is realized by rotation method in the queue, and the balance operation is realized based on priority order outside the queue. Although this method has high processing accuracy, it has the defect of low efficiency. In order to improve the load balancing of electronic communication system, a shunt controller of electronic communication system is designed and implemented based on dynamic Hash algorithm. The shunt controller can balance the shunt operation of router and optimize the load balancing of electronic communication network. The load balancing algorithm based on dynamic Hash algorithm can shape the shunt controller in the electronic communication system, and manage the shunt controller by means of software detection. When the shunt behavior occurs, the strategy shunt is implemented by dynamic Hsah algorithm to ensure that each router can complete the IP message shunt in its own shunt state and realize the load balancing of the electronic communication system. The overall structure of the shunt controller system is shown in Figure 1. The shunt controller uses Telnet component to realize remote login and configure router. It collects all flow groups and interface traffic in router through SNMP component, adjusts local variables, debugs flow groups of all interfaces according to these variables, feeds command files back to router through TFTP server, ensures output flow balance of all interfaces of router, and finally realizes router interface transport. The goal of flow balance is to ensure the load balance of electronic communication system. The shunt controller completes the shunt control of electronic communication system through three functional modules: dynamic update and equalization module, preprocessing module and equalization shunt module. According to the flow of flow groups and interfaces collected in time, the shunt controller arranges the interfaces in order from large to small based on the flow size.
Each interface can only help the corresponding queue to repair. The flow group information can be stored in the nodes, and the strategy shunt can be realized by using the flow group information. Acquisition flow. The shunt controller analyses the client module requirements of SNMP, feeds back the GET application of SNMP to the router, and collects the flow of the interface and the flow of the flow groups on all interfaces to ensure that all local interfaces can complete the repair of a flow group queue, and adjusts the nodes in the local flow group queue to arrange the flow of all interfaces in order from large to small, so as to successfully complete the follow-up work.
Strategic diversion lays the foundation.
Strategic diversion. The average flow of each interface is calculated by dynamic Hash algorithm, and then the difference between all interfaces and average flow is obtained by average flow. According to this difference, the purpose of orderly sorting is achieved. If the difference between interfaces is higher than 0, the flow groups of the interface should be re-divided. The flow groups should be divided in the order of difference from large to small, and the maximum flow interface should be completed. The flow group needs to be connected with the interface with the smallest flow, and which flow group can be connected with the interface with the smallest flow mainly depends on the size of the flow group with the difference and all interfaces. After a series of operations, the router settings will be sorted out and perfected, and then recorded in the relevant documents, so that repeated allocation. File configuration. The shunt strategy records information about settings into the corresponding document through traffic, which is then sent to the router. In order to meet the specific requirements of users, it is necessary to pre-process IP messages of routers. If useful information is sent in time, it is necessary to delete it and then distribute IP messages equally. For example, thermostatic element if users are required to test HTTP messages independently, they must feedback IP messages with 80 source port or destination port to the processor for effective operation, so some redundant processors can not obtain these messages evenly, so user-defined routing schemes should be run before the implementation of traffic balancing schemes. New router.
When a user first uses a router, the first step is to create a new router, which needs to improve the parameters of router address, type, login password, etc. until the shunt controller merges with other parameters of the router, and after setting the original parameters, the user counts the parameters such as IP address and mask of the router interface, and decides the type of interface used; the second step is to use the shunt controller. According to the balanced shunting strategy of dynamic Hash algorithm, the control of initial router traffic is completed. Open the router. Under the premise that the user has only one router document, according to the setting document of opening the router, the association condition between the router and the router can be shaped according to the setting document of opening the router.
The router document can obtain the interface IP address and mask of the router, and can also obtain the list of IP address of the subsequent router according to the type of interface. After the initialization, the dynamic Hash algorithm is used to start the balancing operation of router traffic. Equipment status acquisition. According to the time interval created by users, the shunt controller can collect the current router status and data information in time, and present the results of the flow adjustment of the router interface in the form of dynamic curves. If there are errors in the interface, relevant operations will be carried out to solve the problem.
Dynamic equilibrium. After the traffic information of the router interface is collected by the shunt controller in time, the dynamic Hash algorithm is needed to complete the traffic calculation and control, so that the traffic of the router interface can meet the user’s standard. If it does not meet the user’s standard, the dynamic Hash algorithm is used to complete the balanced control again.
The CAN bus of the shunt controller is shown in Figure 2. It consists of ISA bus, independent CAN communication controller SJA1000, bus transceiver 82C250 and high-speed photoelectric coupler 6N137. In order to improve the load balancing of electronic communication system, based on dynamic Hash algorithm, multiple routers are controlled to balance the shunt and adjust the shunt strategy so as to achieve better shunt effect and achieve load balancing of the system. Dynamic Hash algorithm divides all IP into different stream groups by using Hash function through two-stage Hash method, and identifies Uione for each stream group. The input value of quadratic dynamic Hash function is Uione, and the result of function operation corresponds to the output interface of router.
By analyzing the flow situation of the output interface and adjusting the dynamic Hash function, the flow group allocated to the high flow interface can be ensured, and the flow of the output interface can be adjusted to the low flow interface to realize the dynamic balance of the flow of the output interface. Hash grouping. Using CRC? 32 as input parameter of Hash packet in router IP message can achieve consistency of traffic of each stream group after packet, and has high balanced shunting effect. Flow update and flow group adjustment. Each flow group saves the current flow through a variable, and adjusts the value of the variable after each flow regulation.
According to the flow of each flow group, the corresponding relationship between the flow group and the interface is managed to ensure that the interfaces tend to be balanced. The strategy of dynamic adjustment. Through the first adaptation method, the demand of the difference of the router interface traffic is analyzed, and the flow control of each flow group in the router is completed. The first adaptive method is to collect adjustable streams based on sequence in order to ensure that the interface flow of the acquisition stream group is less than or equal to the interface flow of the transformation stream group until all the transformation interfaces are completed. All the components of the shunt controller in the designed electronic communication system are described by tree. Based on the firewall (ICF) nodes, the policy routing specification (screen) is modeled. The source IP address, address backmask and source port area are segmented based on the specification, and the case of policy routing specification selection is modeled. The performance of the shunt controller designed in this paper is analyzed through experiments. The experiment compares and analyses the load balancing control effects of the proposed method and the rotation method in three aspects: transmission delay, transmission rate and shunting error of an electronic communication network. The performance of the shunt controller designed by this method and rotation method is analyzed experimentally. From Figure 3, it can be seen that the transmission delay of shunt controller in this method is lower than that of rotation method, and the change is more stable, while the delay value of rotation method is higher and there is a large fluctuation. This shows that this method can improve the operation efficiency of shunt controller, mainly because the shunt strategy in this paper can make use of routing resources from a smaller degree to achieve the optimization of network load balancing, and improve the efficiency of shunt controller.
Transmission efficiency. As shown in Figure 4, from the average transmission rate of electronic communication network under different shunting algorithms, it can be seen that the transmission rate of this method is better than that of rotation method. The electronic communication system under shunting controller designed by this method has higher data transmission efficiency. Fig. 5 shows that compared with the rotation method, this method has smaller shunt error, and can accurately schedule the traffic of electronic communication network to achieve network load balancing. This is because the dynamic Hash algorithm proposed in this paper ensures the dynamic balance of the output interface by two-stage Hash method and managing the dynamic Hash function according to the message traffic of the interface.