Allocating optimized resources in the cloud by a viable SLA model