Cloud Task Scheduling Simulation via Improved Ant Colony Optimization Algorithm

As a distributed parallel computing, cloud computing has an absolute advantage in accessing and processing of huge amount of data. How to assign all these virtual cloud computing resources to the user is a key technical issues, scholars have proposed greedy algorithm, FCFS, and other variety of algorithms to solve this problem. However, the algorithms just build a local optimal solution, there is no global allocation of resources in the cloud computing. In order to solve this problem, we are inspired from the ant foraging behavior and integrated multi-goal match of the secondary distribution, then, we proposed ant colony optimization algorithm. The improved ant colony algorithm combines the traditional ant colony algorithm and the greedy algorithm's advantage. Finally, we use the cloud computing simulation tools CloudSim to do the simulative experiments and embed these scheduling algorithms into a visualization system. From the comparison of the simulation results, it can be seen that the improved ant colony algorithm shorten the scheduling time, implement system load balancing at the same time.