Desinging Efficient Matrix Transposition on Various Interconnection Networks Using Tensor Product Formulation

• Main Authors: Chin-Yi Tsai, 蔡進義
• Other Authors: Chua-Huang Huang
• Format: Others
• Language: en_US
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/vsp4re

碩士 === 逢甲大學 === 資訊工程所 === 91 === Matrix transposition is a simple, but an important computational problem. It explores many key issues on data locality. In this paper, we will design matrix transposition algorithms on various interconnection networks, including omega, baseline and hypercube networks. Since different interconnection networks have their own architectural characteristics and properties, an algorithm needs to be tuned in order to be efficiently implemented on various networks. We use a tensor product based algebraic theory to design matrix transposition algorithms on various interconnection networks. This method can be employed to design other block recursive algorithms. A major goal of this study is to provide an effective way for designing VLSI circuits of DSP algorithms.