由於處理器速度與記憶體存取效能的差距越來越大，為了改善記憶存取所帶來的效能降低，可以藉由改善資料區域性來減少主記憶體區塊存取次數。在我們的研究中，我們針對迴圈的資料區域性改善，採用反覆式貪婪演算法，在不違背資料依存性的前提下，反覆的進行強連通元件的搬移來進行迴圈分裂與迴圈融合，並評量搬移後的資料區域性改變，判斷是否接受這次移動。因而可使得循序迴圈的資料區域性逐漸達到最佳化。

The gap between processor speed and memory access performance increases continuingly. In order to improve performance degradation due to memory accesses, we can reduce memory block accesses by improving data locality. In this research, we focus on optimizing data locality in loops. We adopt an iterative greedy algorithm. It iteratively perform loop fission and loop fusion by moving strongly-connected components among loops subject to data dependence constraints. It evaluates data locality change of each move and decide if accept such move. Subsequently loops can be improved gradually forward optimal result.

一章 序論 1
1.1. 研究動機 ..1
1.2. 研究背景 ..2
1.3. 問題敘述 ..4
1.4. 論文組織 ..5
第二章 循序迴圈最佳化方法 6
2.1. 設計環境 ..6
2.2. 迴圈代表法 ..7
2.3. 迴圈轉換 ....14
2.4. 最佳化演算法 17
2.5. 評量方法與模型 19
第三章 循序迴圈最佳化方法的軟體規劃 ...23
3.1. 軟體設計規劃 23
3.2. 迴圈代表法與迴圈轉換軟體設計 24
3.3. 最佳化演算法軟體設計 29
3.4. 評量軟體設計 31
3.5. 評量結果 34
第四章結論 .37
附錄
A、測試程式表 ...38
參考文獻 .46

[1] A. Fraboulet, K. Godary, and A. Mignotte, “Loop Fusion for Memory Space
Optimization,” Proceedings of the International Symposium on System
Synthesis, 2001.
[2] K. Kennedy and K. S. McKinley , “Maximizing Loop Parallelism and
Improving Data Locality via Loop Fusion and Distribution,” Proceedings
of the 6th International Workshop on Languages and Compilers for Parallel
Computing, 1994.
[3] JJ Thompson Avenue, “ Optimizing Inter-Nest Data Locality Using Loop
Splitting and Reordering,” Parallel and Distributed Processing Symposium,
2007.
[4] Tsung-Chuan Huang; Cheng-Ming Yang, “Further results for improving
loop interchange in non-adjacent andimperfectly nested loops,” High
-Level Parallel Programming Models and Supportive Environments, 1998.
[5] K. McKinley , S. Carr, and C. Tseng , “Improving Data Locality with
Loop Transformations,” ACM Transactions on Programming and Systems,
1996
[6] M. Kandemir, I. Kadayif, A. Choudhary, and J. Zambreno, “Optimizing Inter-
Nest Data Locality,” Proceedings of the International Conference on
Compilers, Architecture and Synthesis for Embedded Systems, 2002.
[7] Tsung Lee , Loop Fusion/Fission Optimization of Data Locality , Internal
Research Memo, Computing System Lab. , Dept. of Electrical Eng., Nat’l
Sun Yat-Sen univ. , July 2008
[8] Tsung Lee , A Loop Data Locality Estimation Method , Internal Research
Memo , Computing System Lab. , Dept. of Electrical Eng . ,Nat’l Sun
Yat-Sen univ. , July 2008
[9] Benno Büeler , Andreas Enge , Komei Fukuda, “ Exact Volume computation
for Polytopes: A Practical Study,” CiteSeer.IST Scientific Literature Digital
Library, January 1998