Cloud computing has become an attractive computing paradigm in recent years to offer on demand computing resources for users worldwide. Computing resources are delivered in the form of virtual machines. In such a scenario, task scheduling algorithms play an important role to schedule the tasks effectively to achieve reduction in power consumption and makespan with improvement in resource utilization. Many task scheduling algorithms are introduced to improve energy efficiency of data center. In our work, we have proposed and discussed a power aware dependent task scheduling (PADTS) algorithm and compare it with existing ones.