Simulation of Multi-level Cloud Computing Task Balanced Allocation Based on Genetic Ant Colony
Download as PDF
The scale of cloud computing system is getting bigger and bigger, the topological structure is getting more and more complex, and the heterogeneity of resources makes how to effectively schedule cloud computing tasks become a very important research topic in the field of cloud computing. In this paper, the simulation research of multi-level cloud computing task balance allocation based on genetic ant colony is carried out. In this paper, users' requirements for QoS(Quality of Service) are divided into performance requirements and cost requirements. Performance requirements can shorten the time span by improving the computing performance, transmission performance and storage performance of physical resources, while cost requirements can reduce the computing cost by integrating performance requirements and scheduling costs. A GA _ ACO (Genetic Algorithm-Ant Colony Optimization) algorithm is proposed, which encodes these parameters and finds the optimal combination in the process of evolution. It fully combines the feedback mechanism of ACO, the global search ability of GA and the characteristics of fast convergence. The simulation results show that by combining the respective advantages of GA and ACO, the convergence speed is increased by 4. 07%, and the resource load rate of the algorithm decreases by 30. 28% on average compared with ACO. The load balance of the whole system is gradually improved.
Genetic algorithm; Ant colony optimization; Cloud computing; Balance allocation