臺灣博碩士論文加值系統

多核心處理器 (Multi-Core Processor)為目前中央處理器 (CPU) 主要的發展趨勢，但是一般單執行緒 (Single thread) 的程式無法發揮多核心處理器的效能。多執行緒 (Multi-thread) 的程式來可以解決這個問題，不過多執行緒程式的設計卻複雜又耗時，並且不易偵錯。因此，本論文設計並實作一個可供模組化程式開發 (modular programming) 的訊息交換框架 (Message-based Framework － MBF)，以降低多執行緒程式開發的複雜度，並使程式的開發易於多人分工。在本論文中詳細描述MBF的設計方式，以及說明如何利用MBF來開發應用程式。論文最後測試MBF以多執緒模式運作時相較單執緒模式的效能提升，並針對MBF所提供的計時器 (Timer) 做準確度的測試。經由實驗證明，MBF確實能以多執行緒的方式使程式的效能提升 (Speedup) 接近於處理器核心的數目，並且提供了誤差極低的計時功能。

Multi-Core Processor is the current trend of CPU design. However, a traditional single thread program cannot use all cores in a CPU chip at the same time. Using a multi-thread program is a better way to gain more performance at a multi-core processor. An essential issue is that the design process of a multi-thread program is more complex. Thus, we design and implement a message-based framework (MBF) to solve this problem. MBF also provides a modular programming environment for reducing program development time. In this thesis, we introduce MBF and how to use MBF to develop a multi-thread program. Finally, we present the performance of MBF.

第 1 章 簡介 1
1.1 多核心處理器 1
1.2 多執行緒應用程式 3
1.3 模組化程式設計 5
1.4 模組間通訊 6
1.5 MBF 7
第 2 章 MBF軟體架構 9
2.1 使用者模組資料庫 (User Module Database) 11
2.2 多階層訊息佇列 (Multilevel Message Queue) 13
2.3 控制模組 (Control Module) 14
2.3.1 初始化階段 14
2.3.2 等待階段 16
2.3.3 模組運作階段 18
2.4 訊息遞送模組 (Message Delivery Module) 20
2.5 網路模組 (Network Module) 22
2.5.1 伺服器功能 23
2.5.2 客戶端功能 25
2.6 計時器模組 (Timer Module) 29
第 3 章 MBF應用程式 32
3.1 訊息結構 32
3.2 使用者模組 33
3.3 MBF運作程序 34
3.4 實例說明 40
第 4 章 效能分析 45
4.1 多執行緒效能提升測試 45
4.2 計時器模組準確度測試 51
第 5 章 結論 53
參考文獻 54
附錄 A MBF函式原型 55
A.1 MBF_Init ( ) 55
A.2 TM_Enable ( ) 55
A.3 HostReg ( ) 55
A.4 ModuleReg ( ) 56
A.5 TCP/UDP/SCTP_Server_Start ( ) 57
A.6 SendMsg ( ) 57
A.7 MBF_Wait ( ) 58
A.8 StartTimer ( ) 58
A.9 StopTimer ( ) 59

[1] Krste Asanovic et al., "The Landscape of Parallel Computing Research: A View from Berkeley," EECS Department University of California, Berkeley, Technical Report No. UCB/EECS-2006-183, Dec. 2006.
[2] Shekhar Borkar, "Design challenges of technology scaling," Micro, IEEE, July-August 1999.
[3] Win. A. Wulf and Sally A. McKee, "Hitting the Memory Wall Implications of the Obvious," ACM Computer Architecture News, vol. 23, no. 1, pp. 20-24, Mar. 1995.
[4] John L. Hennessy and David A. Patterson, Computer Architecture: A Quantitative Approach, 4th ed. San Francisco, CA: Morgan Kaufmann Publishers, 2007.
[5] Shekhar Borkar, "Thousand core chips: a technology perspective," in Proceeding DAC '07 Proceedings of the 44th annual Design Automation Conference, 2007, pp. 746-749.
[6] ARM Ltd. ARM11MPCore Processor. [Online]. http://www.arm.com/products/processors/classic/arm11/arm11-mpcore.php

[7] ARM Ltd. ARM11 MPCore™ Processor Technical Reference Manual Revision: r2p0. [Online]. http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0360f/index.html

[8] Abraham Silberschatz, Peter Baer Galvin, and Greg Gagne, Operating System Concept, 7th ed.: John Wiley & Sons, Inc.
[9] Sato Mitsuhisa, "OpenMP: parallel programming API for shared memory multiprocessors and on-chip multiprocessors," in Proceeding ISSS '02 Proceedings of the 15th international symposium on System Synthesis, 2002, pp. 109-111.
[10] Roger S. Pressman, Software Engineering – A Practitioner's Approach, 5th ed.: McGraw-Hill.
[11] Yi-Bing Lin and Ai-Chun Pang, Wireless and Mobile All-IP Networks. Crosspoint Boulevard, Indianapolis: Wiley Publishing, Inc., 2005.
[12] Daniel P. Bovet and Marco Cesati, Understanding the Linux Kernel, 3rd ed.: O'Reilly Media, Inc., 2005.