Ye Yilong

Hangzhou Zhi Ke Fei Chuang Information Technology Co., Ltd.

Abstract：

Against the background of rapid development of big data and cloud computing, resource scheduling has become a key factor restricting the operating efficiency of cloud platforms. The heterogeneous distribution of underlying resources, diverse task demands, dynamic business fluctuations, and high energy consumption of data centers pose severe challenges to traditional scheduling strategies. This paper analyzes the core dilemmas of cloud computing resource scheduling in big data environments from three aspects: resource heterogeneity, task diversity, and system dynamism. It introduces four typical intelligent optimization algorithms—genetic algorithm, particle swarm optimization, ant colony algorithm, and simulated annealing algorithm—and summarizes their improved mechanisms and application effects in heterogeneous and dynamic cloud cluster scheduling. On this basis, four targeted optimization strategies are proposed: multi-objective collaborative optimization, dynamic adaptive adjustment, unified resource integration, and green energy consumption optimization. Practical application tests show that the optimized intelligent algorithms can significantly improve resource utilization, reduce task delay and timeout rate, enhance cluster stability, and lower data center energy consumption.

Key Words：

big data environment; cloud computing; resource scheduling; intelligent optimization algorithm; multi objective optimization; dynamic scheduling