An Agent-Based Emergent Task Allocation Algorithm in Clouds
Chao Chen, Xiaomin Zhu, Weidong Bao, Lidong Chen, Kwang Mong Sim · 2013
Cloud computing has become a promising platform for dealing with emergent tasks. Meanwhile, dynamic task allocation algorithm plays a very important role in obtaining high performance computational capabilities. Unfortunately, little work has been done for emergent task scheduling under Cloud computing environment. To address this issue, we put forward a novel agent-based allocation algorithm (ABAA for short). The algorithm employed the fair competition principle of a roulette to accomplish load balancing, and adopted the dynamic adjustment principle of a buffer pool to accommodate the diversification of task arrival. We conducted extensive experiments on CloudSim platform to evaluate the performance of the strategy. The experimental results indicate that the proposed algorithm can efficiently solve emergent task allocation problem in Clouds.