Optimization For Simulation Task Scheduling On The Integrated Test Management Platform

With the continuous development of computer simulation technology, simulation test has gained more and more people's attention. Simulation test can reduce the cost of testing and reduce the risk of test as well. With the constant expansion of the scale of simulation tests, using only one or a few computers to execute simulation tests has been far from the requirements of satisfying the computing power and simulation test data processing. In order to meet the needs of the simulation tests, distributed simulation technology with HLA standard has been widely used in many fields. However, HLA simulation is only concerned about the high level of the simulation grid, for example, the time synchronization of various components of the simulation and communication mechanisms among them. HLA simulation lacks the effective management of distributed computing resources and scheduling, resulting in substantial waste of resources.In order to solve the above problems which exists in the HLA simulation system, this paper designed and implemented an integrated test management platform middleware. The integrated test management platform middleware provides a unified user interface. users can log on to the user interface to operate the simulation tasks and computing resources. Meanwhile, the resource scheduling and power management strategy has also added in order to automatically monitor simulation tasks and computing resources and manage them. This article migrates the grid resource scheduling and power management into the network middleware, builds a service in the distributed simulation system integration test management platform network middleware. Based on the integrated test management platform middleware, this paper focuses on the resource scheduling and power management issues. Based on the heuristic algorithm Ant Colony Algorithm, the improved Ant Colony Algorithm is proposed and implemented. The aim is scheduling the currently running simulation tasks to the reasonable computing resources so as to achieve the purpose of reducing the power of grid simulation. Based on the dynamic trend prediction algorithm, this paper predicts the load of the simulation grid reasonably. The purpose is managing the computers in simulation grid, so as to achieve the purpose of resource reservation and reduce the energy in the simulation grid.The simulation results show that the simulation grid resource scheduling and power management strategies can effectively schedule the distributed simulation tasks to the reasonable computing resources, and reduce the power consumption in the simulation grid.