Task deployment system for distributed simulation of computer network attack and defense training

Abstract

The invention discloses a task deployment system for the distributed simulation of computer network attack and defense training. In the system, a performance measurement unit is used to calculate the available memory mems of simulation equipment, the main frequency fs of a CPU and the average utilization rate us of the CPU and measure storage performance, an available clock is extracted according to a clock-CPU relationship expressed by cpus=fsXt0X(1-us), and ant colony-atom-task matching is used to process the available clock cpus, the available memory mems and the atom task-simulation relationship information to obtain a simulation sub task. The task deployment system for simulation adopts an ant colony-based task matching algorithm, so a preferable distribution scheme can be selected; experiments show that the distributed attack and defense simulation deployed by the system can reduce actual operation time by 18.3298 percent; the number of attack and defense events supported by the distributed simulation deployed by the system is improved by 40 to 50 percent compared with that before the deployment.

Description

technical field [0001] The invention relates to a computer network security simulation, more particularly, a system for distributed simulation and task deployment for computer network attack and defense drills. Background technique [0002] Today, research on large-scale networks mainly uses modeling and simulation (M&S) technology. Through modeling and simulation technology, it is possible to study and evaluate the performance of network protocols and algorithms under large-scale networks, as well as study large-scale network security incidents. However, building a large-scale network model in network simulation not only requires a lot of computing and storage resources, but also takes a long time to run the simulation. In order to solve these problems caused by large-scale network security simulation, most network simulators currently use Parallel and Distributed Event Simulation (PDES) technology, and use distributed multi-machine systems to jointly complete simulation t...

AI Technical Summary

Problems solved by technology

To solve this problem, the simulation task can be divided among the simulators by using the graph partition method to balance the cooperation between the simulators; the load balancing task decomposition method can also be used to complete the task according to the amount of calculation and communication required. However, the former has the problem of large amount of calculation and local optimality, and the latter has the problem of not considering the relationship between attack and defense, which may lead to unrealistic network simulation.

Images