大量資料需求的分散運算為現今不可或缺的運算方式，如何有效運用網路頻寬也越顯得重要。
　　在過去，我們經常把資料分散在不同的主機上，需要存取時再把資料取回運算。不難想像，我們需要的資料量越是大量，網路的資料的傳輸量也越會隨之增加。
　　所以，在資料可以分類並各類資料數量平均的前提下，我們可以直接將運算向資料儲存節點遷移。在Open MPI的各計算節點上，我們設計令其佈建各類資料，讓Open MPI節點運算時直接使用當地資料，而非透過網路傳輸來獲取資料，如此便將可大幅減少網路的資料傳輸量，有效的利用網路頻寬。

Nowadays, It is indispensable that the amount of transfer data in distribute computing is large. How to use network bandwidth effectively is more and more important.
　In the past, we put the data in many different hosts. If we need the data, we access them through network transfer. It is not hard to image that if we need more and more data, the amount of network traffic is more and more increased.
　Therefore, under the premise that data can be classified and distributed, we move computing into the data node. On every Open MPI node, we design a lot of data which can be classified, and then we put those data to different data nodes. As a result, the computing on Open MPI node can use local data directly without using the data through network transfer. So, we can reduce the amount of data on network and use bandwidth effectively.

致謝 i
摘要 ii
Abstract iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 2
1.3 研究方法與步驟 3
1.3.1 研究方法 3
1.3.2 研究步驟 5
1.4 章節介紹 6
第二章 研究相關介紹 7
2.1 雲端運算介紹 7
2.2 MPI簡介 8
2.3 Open MPI簡介 9
2.4 Sysstat簡介 14
2.5 Memcached簡介 15
2.6 Redis簡介 19
第三章 系統設計與實作說明 21
3.1 系統資料 21
3.2 系統架構設計 24
3.2.1 系統簡介 24
3.2.2 系統架構 26
3.2.2.1 資料儲存 26
3.2.2.2 分派工作與運算 27
3.2.2.3 運算交易策略介紹 32
3.2.2.3.1 策略一 33
3.2.2.3.2 策略二 35
3.2.2.3.3 策略三 37
3.2.2.3.4 策略四 39
3.2.2.3.5 策略五 41
3.2.2.4 工作排程 44
第四章 效能分析測試比較 49
4.1 與分散式儲存節點架構的比較 49
4.2 比較結果 50
4.2.1 運算效能比較 50
4.2.2 網路傳輸量比較 54
第五章 結論與未來展望 58
Reference 59
Appendix 62

[1] Gianni Fenu and Simone Surcis, “A Cloud Computing based Real Time Financial System”, Eighth International Conference on Networks, 2009.
[2] Aiwu Shi, Youfu Xia and Haiyan Zhan, “Applying Cloud Computing in Financial Service Industry”, International Conference on Intelligent Control and Information Processing, 2010.
[3] “Program Trading Continues To Increase”, Elkin McSherry LLC, October, 2005.
[4] “SEC runs eye over high-speed trading”, Financial Times, July 29, 2009.
[5] 姜林杰祐,“程式交易：觀念、方法、技術與解決方案”, 新陸書局出版, 2009.
[6] “OpenMPI”, http://www.open-mpi.org/
[7] “Memcached”, http://memcached.org/
[8] “Redis”, http://redis.io/
[9] “SYSSTAT”, http://sebastien.godard.pagesperso-orange.fr/
[10] 王鵬, “雲端運算的關鍵技術與應用實例”, 佳魁資訊出版, 2010.
[11] “Message Passing Interface”, http://www.mcs.anl.gov/research/projects/mpi/
[12] “MPI”, http://en.wikipedia.org/wiki/Message_Passing_Interface
[13] Jeffrey Dean and Sanjay Ghemawat, “MapReduce: simplified data processing on large clusters”, January 2008. 
[14] Edgar Gabriel, Graham E. Fagg, George Bosilca, Thara Angskun, Jack J. Dongarra, Jeffrey M. Squyres, Vishal Sahay, Prabhanjan Kambadur, Brian Barrett, Andrew Lumsdaine, Ralph H. Castain, David J. Daniel, Richard L. Graham and Timothy S. Woodall, “Open MPI: Goals, Concept, and Design of a Next Generation MPI Implementation”, September, 2004.
[15] “Open MPI v1.6.4 documentation”, http://www.open-mpi.org/doc/v1.6/
[16] “iostat(1) ”, http://sebastien.godard.pagesperso-orange.fr/man_iostat.html
[17] “mpstat (1) ”, http://sebastien.godard.pagesperso-orange.fr/man_mpstat.html
[18] “sar (1) ”, http://sebastien.godard.pagesperso-orange.fr/man_sar.html
[19] Brad Fitzpatrick, “Distributed caching with memcached”, August, 2004.
[20] “LiveJournal”, https://zh.wikipedia.org/wiki/LiveJournal
[21] Jeff Bonwick, “The Slab Allocator: An Object-Caching Kernel Memory Allocator”, summer, 1994.
[22] MEMCACHED CONTRIBUTORS, “memcached - protocol”, 2007.
[23] Karl Seguin, “The Little Redis Book”, January, 2012.
[24] Salvatore Ferragamo, “A fifteen minute introduction to Redis data types”.
[25] 姜林杰祐, “程式交易：觀念、方法、技術與解決方案”, 新陸書局出版, 2009, p.249
[26] 姜林杰祐, “程式交易：觀念、方法、技術與解決方案”, 新陸書局出版, 2009, p.251
[27] XQ全球贏家, http://new.xq.com.tw/categories/4
[28] CMoney雲端控股, http://www.cmoney.tw/notice/
[29] Michael L. Pinedo, “Scheduling: theory, algorithms, and systems Forth Edition”, 2012. 
[30] “Scheduling (computing) ”, https://en.wikipedia.org/wiki/Scheduling_ (computing)#Scheduling_disciplines
[31] W.Richard Stevens, “TCP/IP Illustrated, Volume 1: The Protocols”, March, 2011.
[32] W. Stevens, “TCP Slow Start, Congestion Avoidance, Fast Retransmit, and Fast Recovery Algorithms”, January 1997.