隨著這幾年智慧型手機快速地普及，人們對於行動裝置的需求除了具備上網能力之外，在運算能力與電池續航力都有相當高的要求；而ARM 處理器，具備低功耗與低發熱的優點，並且這幾年ARM 處理器在運算速度大幅提升且持續往多核心架構發展下，因此廣泛地出現在各種移動裝置上，比如智慧型手機、平板電腦、機上盒、車載裝置以及感測網路節點內。

隨著雲端運算的興起，服務供應商有建置大型網路數據中心的需求，這些大型數據中心不僅在運作時帶來相當大能源的消耗，也需要額外透過風扇、空調等方式冷卻數量龐大的處理器，所需要冷卻系統的能源消耗也不容輕忽。因此，有效降低數據中心的能源消耗，已經成為目前數據中心設計時，必須解決的主要議題。

本研究於DevKit8000硬體平台之上建置Ubuntu嵌入式作業系統、安裝Java Runtime Environment、移植Hadoop開放平台，提供低耗能平台運行分散式運算的測試環境。藉由研究數據分析，本平台透過Hadoop MapReduce框架可以快速地進行分散式運算，隨著節點增加，資料處理能力達倍數成長。與傳統x86架構平台 (IntelR Atom™ N270)相比，減低45.5%的能源消耗，減少22.6%的處理時間，來完成相同的工作。實現低耗能、高效率的雲端運算平台之目標。

As the rapid popularization of smart phone in recent years, the users request from mobility device not only for accessing the internet, but also for high computing capability and the sustainability of battery. Under the tendency of dramatic improvement of computing speed and the development of multi-core structure, ARM (Advanced RISC Machine) processors, which carry advantages of low-power consumption and high battery endurance, are applied extensively to various mobility devices such as smart phone, table PC, set-up box, in-car devices and a node of the sensor network.

With the rise of the cloud computing, the service providers establish large Internet Data Center is essential. These Internet Data Centers not only consume great energy for operation and also need to cool down a large number of processors by fan or air conditioner; the quantity of such additional energy required by the cooling system should be put into consideration when designing the IDC. Thus, the power consuming of IDC has become the main issue in coming future.

This study is to provide a solution of decreasing the power consuming by establishing Ubuntu embedded operating system, installing Java Runtime Environment, and transplant Hadoop on the hardware platform of DevKit8000 in order to create a low-power consuming testing environment. According to the experimental data, the designed platform of the study is able to fasten the speed of distributed computing. When the amount of nodes is increase, the capability of data processing grows as a constant percentage. Comparing with the traditional x86 platform of IntelR Atom™ N270, it consumes less energy to accomplish the same task by reducing energy of 45.5% and shortening the duration by 22.6% to achieve the goal of low power, high performance cloud computing platform.

[1] M. Chen, H. Zhang, Y.Y. Su, X. Wang, G. Jiang, K. Yoshihira, “Effective VM sizing in virtualized data centers,” IFIP/IEEE International Symposium on Integrated Network Management, pp. 594-601, 2011.
[2] Cortex™-A15, http://www.arm.com/zh/products/processors/cortex-a/cortex-a15.php (last
visited May 9, 2012)
[3] M. Black and W. Edgar, “Exploring Mobile Devices as Grid Resources: Using an x86 Virtual Machine to Run BOINC on an iPhone,” IEEE/ACM International Conference on Grid Computing, pp. 9-16, 2009.
[4] Apache Hadoop, http://hadoop.apache.org (last visited May 1, 2012)
[5] DevKit8000, http://elinux.org/DevKit8000 (last visited April 20, 2012)
[6] T.G. Wu, J.H. Zhou and J.J. Pan, “A Research of DCT Algorithm Based on OMAP3530,” Proceeding of the International Conference on Wireless Communications Networking and Mobile Computing, pp.1-4, 2010.
[7] Ubuntu Wiki, http://zh.wikipedia.org/wiki/Ubuntu (last visited April 4, 2012)
[8] GNOME, http://www.gnome.org (last visited March 25, 2012)
[9] ARM Ubuntu, http://www.ubuntu.com/download/arm (last visited April 20, 2012)
[10] HDFS™, http://hadoop.apache.org/hdfs/ (last visited May 1, 2012)
[11] Hadoop™ MapReduce, http://hadoop.apache.org/mapreduce/ (last visited May 1, 2012)
[12] DevKit8000 Ubuntu, http://elinux.org/Devkit8000_Ubuntu (last visited April 10, 2012)
[13] DevKit8000 MAC Address, http://devkit8000.wikispaces.com/MAC+Addresses (last visited April 20, 2012)
[14] JavaSE Embedded, http://www.oracle.com/technetwork/java/embedded/downloads/javase/
index.html (last visited April 20, 2012)
[15] Xen ARM, http://xen.org/products/xen_arm.html (last visited June 9, 2012)
[16] Xen ARM Wiki, http://wiki.xen.org/wiki/XenARM (last visited June 9, 2012)
[17] Texas Instruments, http://www.ti.com/ (last visited May 1, 2012)
[18] Texas Instruments. (2010, April). OMAP35x Application Processor: Technical Reference Manual (Literature No. SPRUF980). Dallas, Texas.
[19] D.P. Pham, C.F. Lin, S.M. Yuan and E. Jou, “Database Backed by Cloud Data Store for On-premise Applications,” IEEE 13th International Conference on High Performance Computing and Communications (HPCC), pp. 708-713, September 2011.
[20] J. Xie, S. Yin, X.J. Ruan, Z.Y. Ding, Y. Tian, J. Majors, A. Manzanares and X. Qin, “Improving MapReduce Performance through Data Placement in Heterogeneous Hadoop Clusters,” IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum, pp.1-9, April 2010.
[21] M. Bhandarkar, “MapReduce Programming with Apache Hadoop,” In Proceedings of IEEE International Symposium on Parallel & Distributed Processing, pp. 1, April 2010.
[22] S. Rozsnyai, A. Slominski, Y. Doganata, “Large-Scale Distributed Storage System for Business Provenace,” IEEE International Conference on Cloud Computing, pp.516-524, July 2011.
[23] L. Yang and Z.Z. Shi, “An Efficient Data Mining Framework on Hadoop using Java Persistence API,” IEEE 10th International Conference on Computer and Information Technology (CIT), pp. 203-209, 2010.
[24] R.T. Kaushik, M. Bhandarkar and K. Nahrstedt, “Evaluation and Analysis of GreenHDFS: A Self-Adaptive, Energy-Conserving Variant of the Hadoop Distributed File System,” IEEE Second International Conference on Cloud Computing Technology and Science (CloudCom), pp. 274-287, 2010.
[25] G. Kousiouris, G. Vafiadis and T. Varvarigou, “A Front-end, Hadoop-based Data Management Service for efficient Federated Clouds,” IEEE Third International Conference on Cloud Computing Technology and Science (CloudCom), pp. 511-516, 2011.
[26] S. Sathya and M.V. Jose, “Application of Hadoop MapReduce technique to Virtual Database system design,” International Conference on Emerging Trends in Electrical and Computer Technology (ICETECT), pp. 892-896, 2011.
[27] J. Ma and J.J. Lu, “The architecture of Tender and Bidding System of enterprises based on Hadoop Cloud Platform,” International Conference on Electric Information and Control Engineering (ICEICE), pp. 6234-6237, 2011.
[28] D.W. Zhang, F.Q. Sun, X. Cheng and C. Liu, “Research on hadoop-based enterprise file cloud storage system,” 3th International Conference on Awareness Science and Technology (iCAST), pp. 434-437, 2011.
[29] Y.F. Li, W.Q. Li and C.F. Jiang, “A Survey of Virtual Machine System: Current Technology and Future Trends,” 3th International Symposium on Electronic Commerce and Security (ISECS), pp. 332-336, 2010.
[30] N. Kim, J. Cho and E. Seo, “Energy-Based Accounting and Scheduling of Virtual Machines in a Cloud System,” IEEE/ACM International Conference on Green Computing and Communications (GreenCom), pp. 176-181, 2011.