Recently, the power consumption of data centers has been increasing due to the advance of digitalization across various industries and the proliferation of cloud computing. While there are technologies that reduce power consumption at the expense of application performance, this trade-off presents a significant challenge for carrier-grade applications that must meet stringent reliability and availability requirements. This research aims to further reduce power consumption and costs in cloud environments while ensuring the responsiveness of carrier-grade applications. We have defined three requirements to achieve a carrier-grade power-saving system: application responsiveness, resource efficiency, and power efficiency. We propose a cloud system that meets those three requirements in use cases where applications with redundancy operate in a cloud environment. This is achieved by abstracting the redundancy protocols specific to each application and defining and introducing a new operational interface for redundancy switching. The proposed system features a unique set-packing algorithm that determines the optimal redundancy configuration to maximize power consumption efficiency, enabling autonomous reconfiguration. An evaluation of a simplified implementation of this system revealed it improved power efficiency by 18% compared to conventional systems.