Abstract

Although technique known as server consolidation approach in a data center can reduce the overall power consumption, the Power Usage Effectiveness (PUE) of the data center will still be negatively affected with presence of distributed Uninterruptible Power Supplies (UPSs). The impact on the PUE arises from the fact that all UPS modules are kept running to maintain power availability for only a few active servers during off-peak periods. To address this problem, in this paper technique for reducing power consumption in a data center by consolidating the UPSs used during off peak periods is proposed. The proposed technique achieves power savings by leveraging a micro Automatic Transfer Switch (micro-ATS) at the server end. The novelty of this work lies in developed adaptive algorithm that continuously looks for opportunities to reduce the number of UPSs by offloading under-loaded UPSs to a neighboring UPS whenever that neighboring UPS can handle the extra load. In various simulated scenarios involving corporate data centers, our approach demonstrates the ability to save more power and achieve lower PUE degradation compared with state-of-the-art approaches such as server consolidation. Specifically, the proposed approach achieves a savings of approximately 20% to 40% in a data center’s power consumption, depending on the data center’s off-peak periods, which can be accomplished using only 80% of the UPS modules in the data center.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.