Reputation guided genetic scheduling algorithm for independent tasks in inter-clouds environments

Evolutionary computing offers different methods to solve NP-hard problems, finding a near-optimal solution. Task scheduling is a complex problem for large environments like Clouds. Genetic algorithms are a good method to find a solution for this problem considering multi-criteria constrains. This is also a method used for optimization. In these type of environments service providers want to increase the profit and the customers (end-users) want to minimize the costs. So, its all about money and we have minimum two optimization constrains. On the other hand, a good technique to ensure the QoS is to use the reputation of resources offered. This aspect is very important for service providers because represents a ranking method for them. We present in this paper a reputation guided genetic scheduling algorithm for independent tasks in inter-Clouds environments. The reputation is considered in the selection phase of genetic algorithm as an evolutionary criteria for the algorithm. We evaluate the proposed solution considering load-balancing as a way to measure the optimization impact for providers and maxspan as a metric for user performance.