本研究之主要目的是設計與實現基於ARM為基礎具多核心與多中央處理器之嵌入式伺服器系統工作板(ARM-Based Multi-Cores & Multi-CPUs Server Board, M2SB)，以提升ARM叢集式系統的整體效能，實現高能源效率之伺服器系統設計。
在嵌入式系統的市場中，ARM架構基於低功率消耗的特性，其應用領域已逐漸擴展到伺服器架構。文獻回顧顯示：已有許多研究著重於ARM架構的效能分析研究，並也將這些結果拿來和已做過效能最佳化的x86系統來做比較。然而，根據以往的實驗數據，並不足以證明，ARM處理器在伺服器系統上的應用也可以得到一些優勢。因此本研究因應企業需求採用伺服器級的ARM四核心處理器設計，以作相關研究與驗證。在本設計中，主要提供PCI Express和Gigabit等級的乙太網路兩種型態的傳輸介面，這樣的設計將使ARM伺服器系統能和其他系統作溝通。最後，再透過平行運算標準和伺服器工作負載兩個實驗，可以證明M2SB具有低耗能的優點，驗證了ARM處理器在伺服器系統應用上的優勢。

The objective of this research is to design and implement an ARM-Based Multi-Cores & Multi-CPUs Server Board (M2SB). Such design not only can improve the entire performance of ARM cluster systems but also can acheive the high-energy-efficiency server system implementation.
In the embedded systems’ market, due to the benefit of the low power consumption in the ARM architecture, the related ARM applications have been expanded to the server architecture. Literature survey indicates that there are many existing ARM-based performance analysis, which have been compared with optimized x86 systems. However, according to the former experimental data, we cannot ensure if ARM-based server systems will get better performance. Therefore, in this study, we adopt the server-grade ARM quad-core processors for enterprise-class applications. Various types of cluster systems can be built with the M2SB by either the PCI Express interface or the Gigabit Ethernet interface. Finally, through the experiments of parallel workloads and server workloads, we can demonstrate that M2SB owns the benefit of low power consumption and clearly verify that in server systems ARM processors will be able to work with better performance than x86 prosessors do.

[1]Electronic Industries Alliance,“EIA Standards,”http://www.eciaonline.org/eiastandards/.
[2]Mellanox Technologies,“InfiniBand Cards,”http://www.mellanox.com/page/infiniband_cards_overview.
[3]InfiniBand Trade Association,“InfiniBand Architecture Specification,”http://www.infinibandta.org/content/pages.php?pg=technology_download.
[4]PCI-SIG,“PCI Express,”http://www.pcisig.com/home.
[5]“CentOS,”http://www.centos.org/.
[6]陳聰杰，「監控/感測系統相輔相成　自動化控制設備增添智慧功能」，城邦文化事業股份有限公司，新通訊元件雜誌，第154期，December 2013。
[7]Qualcomm Incorporated,“Discover the full line of Snapdragon processors,”http://www.qualcomm.com/snapdragon/processors.
[8]Marvell Technology Group Ltd.,“ARMADA XP,”http://www.marvell.com/embedded-processors/armada-xp/.
[9]IBM Corporate,“The Cell Broadband Engine,”http://www-03.ibm.com/ibm/history/ibm100/us/en/icons/cellengine/.
[10]Intel Corporation,“Intel Xeon Phi Product Family,”http://www.intel.com/content/www/us/en/processors/xeon/xeon-phi-detail.html.
[11]Mercury Computer Systems,“Mercury Computer Systems Brings the Cell BE Processor to the PC Workstation Architecture,”http://www.mrcy.com/search.aspx?searchtext=Mercury%20Computer%20Systems%20Brings%20the%20Cell%20BE%20Processor%20to%20the%20PC%20Workstation%20Architecture.
[12]Calxeda Corporate,“Quad-Node EnergyCard,”http://www.calxeda.com/technology/%20products/energycards/quadnode/.
[13]ARM Ltd.,“The Architecture for the Digital World,”http://www.arm.com/, 2013.
[14]E. L. Padoin, D. A. G. Oliveira, P. Velho, and P. O. A. Navaux,“Time-to-solution and energy-to-solution: A comparison between ARM and Xeon,”Proceedings Third Workshop on Application for Multi-Core Architectures, New York, NY, USA, October 24-25, 2012, pp. 48-53.
[15]P. Stanley-Marbell and V. C. Cabezas,“Performance, Power, and Thermal analysis of low-power processors for scale-out systems,”2011 IEEE International Parallel & Distributed Processing Symposium, Shanghai, China, May 16-20, 2011, pp. 863-870.
[16]C. Landi, P. Merola, and G. Ianniello,“ARM-based energy management system using smart meter and Web server,”2011 IEEE Instrumentation and Measurement Technology Conference, Binjiang, China, May 10-12, 2011, pp. 1-5.
[17]M. Olson, K. Christensen, S. H. Lee, and J. Yun,“Hybrid Web server: Traffic analysis and prototype,”36th Annual IEEE Conference on Local Computer Networks, Bonn, Germany, October 4-7, 2011, pp. 131-134.
[18]M. Mostowfi, K. Christensen, S. H. Lee, and J. Yun,“Timed Redirection: HTTP request coalescing to reduce energy use of hybrid Web servers,”37th Conference on Local Computer Networks, Clearwater, FL, USA, October 22-25, 2012, pp. 168-171.
[19]M. Yamagiwa and M. Uehara,“A study on constructing an energy saving cloud system powered by photovoltaic generation,”2012 15th International Conference on Network-Based Information Systems, Melbourne, VIC, Australia, September 26-28, 2012, pp. 844-848.
[20]G. D. Prete and C. Landi,“Real-time smart meter with embedded web server capability,”2012 IEEE International Instrumentation and Measurement Technology Conference, Graz, Austria, May 13-16, 2012, pp. 682-687.
[21]D. Jiang, M. Tian, W. C. Liu, and Y. C. Pan,“Design and implementation of a novel wireless sensor network system terminal based on embedded web server and database,”International Conference on Automatic Control and Artificial Intelligence, Xiamen, China, March 3-5, 2012, pp. 772-775.
[22]Y. W. Bai, J. P. Hsu, H. Teng, and H. E. Lin,“Design and implementation of an embedded surveillance system with video streaming recording triggered by an infrared sensor circuit,”2007 International Symposium on Communications and Information Technologies, Sydney, Australia, October 17-19, 2007, pp. 335-340.
[23]E. Principi, V. Colagiacomo, S. Squartini, and F. Piazza,“Low power high-performance computingon the Beagleboard platform,”2012 Proceedings of the 5th European DSP Education and Research Conference, Amsterdam, Netherlands, September 13-14, 2012, pp. 35-39.
[24]D. Flynn,“Power gating applied to MP-SoCs for standby-mode power management,”2013 50th ACM / EDAC / IEEE Design Automation Conference, Austin, TX, USA, May 29-June 7, 2013, pp. 1-5.
[25]Dell Inc.,“Copper enables the ARM server ecosystem,”http://www.dell.com/Learn/us/en/555/campaigns/project-copper?c=us&l=en&s=biz.
[26]D. Goddeke, D. Komatitsch, M. Geveler, D. Ribbrock, N. Rajovic, N. Puzovic, and A. Ramirez,“Energy efficiency vs. performance of the numerical solution of PDEs: an application study on a low-power ARM-based cluster,”Journal of Computational Physics, vol. 237, 2013, pp. 132-150.
[27]E. Blem, J. Menon, and K. Sankaralingam,“Power struggles: Revisiting the RISC vs. CISC debate on contemporary ARM and x86 architectures,”Proceedings of 19th IEEE International Symposium on High Performance Computer Architecture (HPCA), Shenzhen, China, February 23-27, 2013, pp. 1-12.
[28]E. L. Padoin, D. A. G. Oliveira, P. Velho, and P. O. A. Navaux,“Evaluating performance and energy on ARM-based clusters for high performance computing,”2012 41st International Conference on Parallel Processing Workshops, Pittsburgh, PA, USA, September 10-13, 2012, pp. 165-172.
[29]PLX Technology,“PCIe Switches,”http://www.plxtech.com/products/expresslane/switches.
[30]K. J. Bowers, B. J. Albright, B. Bergen, and L. Yin,“0.374 Pflop/s trillion-particle kinetic modeling of laser plasma interaction on roadrunner,”SC 2008 Proceedings of the 2008 ACM/IEEE conference on Supercomputing, Austin, TX, USA, November 15-21, 2008, pp. 1-11.
[31]C. H. Hsu and W. C. Feng,“A Power-aware Run-time system for High-performance computing,”Proceedings of the 2005 ACM/IEEE conference on Supercomputing, Seattle, WA, USA, November 12-18, 2005.
[32]J. S. Chase, D. E. Irwin, L. E. Grit, J. D. Moore, and S. E. Sprenkle,“Dynamic virtual clusters in a grid site manager,”Proceedings 12th IEEE International Symposium on High Performance Distributed Computing, Seattle, WA, USA, June 22-24, 2003, pp. 90-100.
[33]A. J. Younge, G. Laszewski, L. Wang, S. Lopez-Alarcon, and W. Carithers,“Efficient resource management for cloud computing environments,”Chicago, IL, USA, August 15-18, 2010, pp. 357-364.
[34]K. J. Barker, K. Davis, A. Hoisie, D. K. Kerbyson, M. Lang, S. Pakin, and J. C. Sancho,“Entering the petaflop era: The architecture and performance of roadrunner,”SC 2008 International Conference for High Performance Computing, Networking, Storage and Analysis, Austin, TX, USA, November 15-21, 2008, pp. 1-11.
[35]G. Laszewski, A. Younge, H. Xi, K. Mahinthakumar, and L. Wang,“Experiment and workflow management using Cyberaide shell,”9th IEEE/ACM International Symposium on Cluster Computing and the Grid, Shanghai, China, May 18-21, 2009, pp. 568-573.
[36]R. Buyya, C. S. Yeo, and S. Venugopal,“Market-oriented cloud computing: Vision, Hype, and Reality for delivering IT services as computing utilities,”The 10th IEEE International Conference on High Performance Computing and Communications, Dalian, China, September 25-27, 2008, pp. 5-13.
[37]G. Laszewski, L. Wang, A. J. Younge, and H. Xi,“Power-aware scheduling of virtual machines in DVFS-enabled clusters,”2009 IEEE International Conference on Cluster Computing, New Orleans, LA, USA, August 31- September 4, 2009, pp. 1-10.
[38]D. J. Kerbyson, H. J. Alme, A. Hoisie, F. Petrini, H. J. Wasserman, and M. Gittings,“Predictive performance and scalability modeling of a Large-scale application,”Proceedings of the 2001 ACM/IEEE conference on Supercomputing, New York, NY, USA, November 10-16, 2001.
[39]F. Petrini, D. K. Kerbyson, and S. Pakin,“The case of the missing supercomputer performance: Achieving optimal performance on the 8,192 processors of ASCI Q,”Proceedings of the 2003 ACM/IEEE conference on Supercomputing, Phoenix, AZ, USA, November 15-21, 2003.
[40]NASA Advanced Supercomputing Division,“NAS Parallel Benchmarks,”http://www.nas.nasa.gov/publications/npb.html
[41]Apache,“HTTP SERVER PROJECT, ”http://httpd.apache.org/.
[42]“The lighttpd web server,”http://www.lighttpd.net/.
[43]Microsoft,“Windows XP,”http://windows.microsoft.com/en-us/windows/windows-help?os=winxp#windows=windows-xp.
[44]Canonical Ltd.,“About Ubuntu,”http://www.canonical.com/about-ubuntu.
[45]Cadence Design Systems, Inc.,“Cadence PCB Design and SI/PI Analysis,”http://www.cadence.com/products/pcb/Pages/pcb_orcad_sigrity.aspx.
[46]Tektronix, Inc.,“Oscilloscope,”http://www.tek.com/oscilloscope#all.
[47]Lauterbach GmbH,“TRACE32,”http://www.lauterbach.com/frames.html?home.html.
[48]Ravi Budruk, Don Anderson, and Tom Shanley, PCI Express System Architecture, MindShare, 2003.
[49]楊志偉，PCI Express 系統架構，碁峰資訊股份有限公司，2007。
[50]Jack Regula, Using Non-Transparent Bridging in PCI Express Systems, White Paper, PLX Technology Inc., June, 2004.
[51]PLX Technology,“Non-Transparent Bridging Simplified Multi-Host System and Intelligent I/O Design with PCI Express,”http://www.plxtech.com/files/pdf/technical/expresslane/NTB_Brief_April-05.pdf, 2004.
[52]Intel Corporation, 21555 Non-Transparent PCI-to-PCI Bridge User Manual, 2001.
[53]Mark I. Montrose, Printed Circuit Board Design Techniques for EMC Compliance-A Handbook for Designers, Second Edition, IEEE Electromagnetic Compatibility Society, Sponsor, Wiley-interscience a John Wiley & Sons, Inc., Publication, 2000.
[54]謝金明，高速數位電路設計暨雜訊防制技術，全華科技圖書股份有限公司，1999。
[55]S. H. Hall, G. W. Hall, and J. A. McCall, High-Speed Digital System Design- A Handbook of Interconnect Theory and Design Practices, John Wiley & Sons, 2000.
[56]Michael Larabel,“ARM on Ubuntu 12.04 LTS battling Intel x86?,”Phoronix Media,http://www.phoronix.com/scan.php?page=article&item=ubuntu_1204_armfeb&num=1.
[57]Z. Ou, B. Pang, Y. Deng, J. K. Nurminen, A. Yla-Jaaski, and P. Hui,“Energy- and cost-efficiency analysis of ARM-based clusters,”2012 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, Ottawa, Canada, May 13-16, 2012, pp. 115-123.
[58]D. Tsirogiannis, S. Harizopoulos, and M. A. Shah,“Analyzing the energy efficiency of a database server,”Proceedings of the 2010 ACM SIGMOD International Conference on Management of data, New York, NY, USA, June 06-11, 2010, pp. 231-242.
[59]Texas Instruments Incorporated.,“OMAP 4 Mobile Applications Platform OMAP4430,”http://www.ti.com/product/omap4430.
[60]NVIDIA Corporation,“Tegra Processors,”http://www.nvidia.com/object/tegra-4-processor.html.
[61]E. Gabriel, G. E. Fagg, G. Bosilca, T. Angskun, J. J. Dongarra, J. M. Squyres, V. Sahay, P. Kambadur, B. Barrett, A. Lumsdaine, R. H. Castain, D. J. Daniel, R. L. Graham, and T. S. Woodall,“Open MPI: Goals, concept, and design of a next generation MPI implementation,”Proceedings of the 11th European PVM/MPI Users’ Group Meeting, Budapest, Hungary, September 19-22, 2004, pp. 97-104.
[62]MVAPICH,“MVAPICH: MPI over InfiniBand, 10GigE/iWARP and RoCE,”http://mvapich.cse.ohio-state.edu/, 2013.
[63]OpenMP Architecture Review Board,“OpenMP Application Program Interface version 3.0,”http://www.openmp.org/mp-documents/spec30.pdf, 2008.
[64]Khronos OpenCL Working Group,“The OpenCL Specification version 1.2,”http://www.khronos.org/registry/cl/specs/opencl-1.2.pdf, 2012.
[65]Janusz Kowalik and Tadeusz Puzniakowski, Using OpenCL: Programming Massively Parallel Computers, ser. Advances in Parallel Computing. IOS Press, 2012.
[66]NVIDIA,“CUDA Parallel Computing Platform,”http://www.nvidia.com/object/cuda_home_new.html.
[67]Semtech Corporation,“Switching Regulators,”http://www.semtech.com/power-management/switching-regulators/.
[68]陳聰杰，「平行運算之電路設計」， 全亞文化事業有限公司，ET電子技術雜誌，第312期，March 2012。
[69]張登章，陳聰杰，徐勝均，「嵌入式系統容錯控制技術架構」，智慧自動化產業期刊，第6期，September 2013。
[70]Carl Hamacher, Zvonko Vranesic, Safwat Zaky, and Naraig Manjikian, Computer Organization and Embedded Systems, Sixth Edition, McGraw-Hill Companies, 2012.