雲端運算技術是近年來熱門的議題，不論是政府機關、民間企業以及研究機構無不專注發展新的技術與應用，Google、Amazon和微軟等大型雲端供應商也在世界各地進行雲端資料中心的建置，這些大型的資料中心也帶來相當大的電力需求以及二氧化碳的排放，由於全球資源的衰竭，電力成本的不斷增加，環保節能的議題也越來越受到重視，雲端運算最需要的資料中心，不論是硬體建置、管理、運作效能、耗電，甚至是設置地點等議題都成為熱門研究的議題。這些大型資料中心裡的伺服器運作上，有相當大的能量損失在於電源供應器供應電力時，AC轉換到DC的功率損失，以及伺服器主機板上直流降壓電路的損耗。
本文以Intel平台以及Lite-on之電源供應器，進行負載與消耗電力之模擬分析，並提出智慧型的電源管理控制方法，透過微處理器監視系統負載狀況，依據負載的高低控制電源供應器的開與關，使得電源供應器操作在轉換效率高的範圍之內，有效的避免系統在輕載時額外的電力損耗。
經過實際量測與分析的結果，電源供應器操作在輕載的狀況時，其電源的轉換效率相當低落，操作於低載時，所浪費的功率很高，使用本文提出的電源控制方法，可以在輕載時提高電源轉換效率最高17.54%，在低載運作時可以減少額外的電力耗損，在日益高漲的電費狀況下，可有效降低資料中心的運作成本。

Recently, cloud computing technology has becoming a popular topic within government agencies, private companies and research institutes. They all focus on development for new cloud computing technologies and applications. Google, Amazon and Microsoft, cloud computing providers, build cloud data centers all around the world. The large data center is considered would bring high demand for electricity and carbon dioxide emissions. Since depletion of global resources, rising costs of electricity and environmental protection, energy saving issues obtain progressive attentions. Extensively there are some state of the art research issues of cloud computing that regardless of the hardware implementation, such as resource management, operational efficiency, power consumption, and even the service location. In cloud computing large data center server operation, energy loss occurs when power supply performs AC to DC conversion likewise the server motherboard DC buck circuit losses.
This study exploits capabilities of Intel platform and Lite-on of power supply to conduct power consumption simulation analysis and finally apply intelligent power management control. It can monitor the system load status through microcontroller, and according to the load level of power supply, the system will be controlled to keep the power supply operation for highly efficient with effective range of power consumption to avoid overload system operation.
By actual measurement and analysis, at light load power supply operation, its power conversion efficiency was very low, on the other hand on low load operation, the wasted power rate is very high. By implementing the proposed power control method power conversion efficiency at light loads can be improved up to 17.54%. Meanwhile, on the low loading operation, the additional power consumption can be reduced. Conclusively, during rising electricity situation, the proposed method can effectively reduce data center operating costs.

[1]EPA Report to Congress on Server and Data Center Energy Efficiency, Public Law 109-431, 2007.
[2]D. Peek and J. Flinn, “Drive-thru: Fast, accurate evaluation of storage power management,” The USENIX Annual Technical Conference, pp.251-264, 2005.
[3]P. Mell and T. Grance, “The NIST Definition of Cloud Computing”, NIST September 2011. http://csrc.nist.gov/publications/nistpubs/800-145/SP800-145.pdf
[4]Cloud computing, http://www.nist.gov/index.html
[5]Google Data Center, http://www.google.com/about/datacenters
[6]PIC16F1829, http://mircochip.com
[7]G. V. Laszewski, L. Wang, A. J. Younge and X. He, “Power-Aware Scheduling of Virtual Machines in DVFS-enabled Clusters,” In Proceedings of the IEEE International Conference, pp. 1-10, 2009.
[8]MPLAB IDE, http://www.microchip.com/pagehandler/en-us/family/mplabx
[9]T. V. Truong, Y. Sato and Y. Inoguchi, “Performance Evaluation of a Green Scheduling Algorithm for Energy Savings in Cloud Computing,” In Proceedings of the Parallel & Distributed Processing Workshops and Phd Forum, IEEE International Symposium, pp. 1-8, 2010.
[10]H. Yuan, J. Kuo and I. Ahmad, “Energy Efficiency in Data Centers and Cloud-Based Multimedia Services: An Overview and Future Directions,” In Proceedings of Green Computing Conference, pp. 375-382, 2010.[
[11]IPMI, http://www.intel.com/design/servers/ipmi
[12]PMBus, http://pmbus.org/index.php
[13]J. Younge, G. V. Laszewski and L. Wang, “Efficient Resource Management for Cloud Computing Environments,” In Proceedings of the IEEE Green Computing Conference, pp. 357-364, 2010.
[14]SPECpower, http://www.spec.org
[15]Y. Pu,Y. Deng and A. Nakao, “Cloud Rack: Enhanced Virtual Topology Migration Approach with Open vSwitch,” In Proceedings of the IEEE Information Networking (ICOIN) International Conference, pp. 160-164, 2011
[16]Xeon E5-2400 series CPU spec. http://ark.intel.com/products/family/59138/Intel-Xeon-Processor-E5-Family/server
[17]Liteon Power supply, http://www.liteon.com/index.aspx
[18]A. Corradi, M. Fanelli and L. Foschini, “Increasing Cloud Power Efficiency through Consolidation Techniques,” In Proceedings of the IEEE Computers and Communications Symposium, pp. 129-134, 2011.
[19]Y. Shi, X. Meng, J. Zhao, X. Hu, B. Liu, and H. Wang, “Benchmarking Cloud-based Data Management Systems,” In Proceedings of the second international worksop on Cloud data management, pp.47-54, 2010.
[20]I. Sarji, C. Ghali, A. Chehab and A. Kayssi,“CloudESE: Energy efficiency model for cloud computing environments,” In Proceedings of the IEEE Energy Aware Computing (ICEAC), International Conferene, pp.1-6, 2011.
[21]J. Lamb, “ Green IT and use of private cloud computing in South Africa,” In Proceedings of the IEEE Emerging Technologies for a Smarter World (CEWIT) 8th International Conference, pp. 1-6, 2011.
[22]BurnInTest, http://www.passmark.com/
[23]Y. Li, H.Zhang and K. H. Kim, “ A Power-Aware Scheduling of MapReduce Applications in the Cloud,” In Proceedings of the IEEE Dependable, Autonomic and Secure Computing (DASC) Conference, pp. 613-620, 2011.
[24]Q. Chen, P. Grosso, K. V. D. Veldt, C. D. Laat, R. Hofman and H. Bal, “Profiling Energy Consumption of VMs for Green Cloud Computing, ” In Proceedings of the IEEE Dependable, Autonomic and Secure Computing (DASC) International Conference, pp. 768-775, 2011.
[25]B. Baikie and L. Hosman, “Green cloud computing in developing regions Moving data and processing closer to the end user,” In Proceedings of the IEEE Telecom World (ITU WT) Technical Symposium, pp. 24-28, 2011.
[26]S. Ricciardi, D. Careglio, U. Fiore and F. Palmieri, “Saving Energy in Data Center Infrastructures,” In Proceedings of the IEEE Data Compression, Communications and Processing (CCP) International Conference, pp.265-270, 2011.
[27]L. Ganesh, H. Weatherspoon and K. Birman, “Beyond Power Proportionality: Designing Power-Lean Cloud Storage,” In Proceedings of the IEEE Network Computing and Applications (NCA) International Symposium, pp.147-154, 2011.
[28]80 PLUS, http://www.plugloadsolutions.com/80PlusPowerSupplies.aspx