頁籤選單縮合
題名 | 圖形學在稀疏矩陣儲存法上之應用=A Graph Theory Application on the Storage of a Sparse Matrix |
---|---|
作者 | 賴泳伶; | 書刊名 | 嘉義師院學報 |
卷期 | 13 1999.11[民88.11] |
頁次 | 頁225-234 |
分類號 | 310.15 |
關鍵詞 | 稀疏矩陣; 圖形; 帶寬; 輪廓; Sparse matrix; Graph; Bandwidth; Profile; |
語文 | 中文(Chinese) |
中文摘要 | 在有限元素方法(finite element method)的運用中,常需要解大量的矩陣線性方程式([M][X]=B),其中矩陣[M]為一個對稱的稀疏矩陣。這類的矩陣也常出現在微分方程或數值分析的係數裡。在電腦的處理速度和儲存能力已大幅提昇之後,解這類的矩陣方程式卻仍需要耗費極長的時間來計算及使用極大的記憶體來儲存。許多人以為圖形學只是純粹的理論,只有學術價值而無法在其他方面有實際的應用。本文特別就圖形上的「帶寬」、「輪廓」等參數在稀疏矩陣儲存法上之應用,詳加敘述。文末並指出這些參數除了可應用在稀疏矩陣的儲存之外,也有其他方面的應用價值。 |
英文摘要 | In the use of the finite element method, it often requires to solve a large set of linear algebraic equations of the form [M][X]=[B] where [M] is a large symmetric sparse matrix. This kind of matrices also appears as coefficient matrices of systems of differential equations in numerical analysis and physics. Although the computer storage capacity and the internal speed has been greatly increased in recent years, the solution of systems of equations still requires a huge amount of memory and computing time. This paper describes the application of the graph parameters-bandwidth and profile on the storage of a sparse matrix. It also points out other applications of those parameters. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。