An Adaptive Simulated Annealing Genetic Algorithm for the Data Placement Problem in Saas

Cloud computing has received a lot of attention and adopted by Software as a Service (SAAS) providers. However, there are still many challenges in placing a SAAS across globally distributed datacenters, such as reducing transmission time and achieve load balancing simultaneously. This paper proposes an adaptive simulated annealing genetic algorithm (ASAGA) approach which can change crossover rate and mutation rate adaptively and combines simulated annealing mechanism to address this problem. Experimental results show that compared with simple genetic algorithm, ASAGA is feasible and scalable, and it has shorter execution time and convergence times.