支援多核心系統架構之多執行緒程式設計環境之建置

BUILDING A MULTI-THREADED PROGRAMMING ENVIRONMENT IN A MULTI-CORE SYSTEM

廖友誌(Yu-Chih Liao)

指導教授 ： 李良德

大同大學/電機資訊學院/資訊工程學系所/碩士(2010年)

摘要

在科技不斷進步的環境下，嵌入式系統的發展已相當成熟，尤其在多媒體應用方面，但是多媒體的應用往往需要大量且複雜的運算，同時系統對於處理器效能的需求也逐漸增加。基於速度與降低功耗的要求，有逐漸往多核心發展的趨勢。在一般電腦系統上，平行程式設計的相關技術已有相當程度的發展，可是嵌入式系統在平行計算上卻缺乏支援性。因此本論文提供一個平行程式架構和函式庫，使程式設計者在多核心嵌入式系統上設計軟體時，有方便的平行化程式設計環境，以縮短程式設計的時間，並有效提升程式平行化後的執行速度。此外，本論文針對平行程式中集體傳訊函式，配合使用的平台加以改良並新增其功能，使得平行程式設計更為方便、執行效能更為提升。

關鍵字

嵌入式系統 ； 平行化程式設計 ； 多核心架構

並列摘要

For the continuously progress of the science and technology, the development of embedded systems is quite mature currently, particularly in multimedia applications. However, multimedia applications often require large and complex computations. While the requirements of the system for processor performance have gradually increased, based on the high computation speed and low power consumption requirements, development of the multi-core is becoming a trend. For the general computer system, parallel programming technology has been developed in a long period, but parallel computing in embedded systems is still lack of supporting. A parallel programming framework with libraries for supporting a multi-core embedded system has been proposed in this thesis. Under the proposed parallel programming environment, programmers can design parallel programs for a multi-core embedded system conveniently. Thus, it can shorten not only the programming time but also the execution time of the parallel program. In addition, collective communication functions for the parallel programming have been improved according to the target platform, and several features have been augmented in the proposed framework to make parallel programming easier. The experimental results show that the performance of the multi-core system can be improved significantly by applying the proposed programming environment for developing parallel program.

並列關鍵字

embedded system ； parallelize programming ； multi-core architecture

參考文獻

[5]Heiko Schwarz, Detlev Marpe, and Thomas Wiegand, “Overview of the Scalable H.264/MPEG4-AVC Extension,” IEEE International Conference on Image Processing, pp. 161-164, Oct. 2006

[6]Li Zhuol, Qiang Wang, David Dagan Feng, and Lansun Shen, “Optimization and Implementation of H.264 Encoder on DSP Platform,” IEEE International Conference on Multimedia and Expo, pp. 232-235, Jul. 2007

[9]“OpenMP Application Program Interface,” available at http://www.openmp.org/

[1]David A. Bader, “Introduction to SWARM Software and Algorithms for Running on Multicore Processors”, available at http://multicore-swarm.sourceforge.net/

[2]David A. Bader, Varun Kanade and Kamesh Madduri, “SWARM: A Parallel Programming Framework for Multicore Processors,” IEEE International Parallel and Distributed Processing Symposium, pp. 1-8, Mar. 2007