Area-time tradeoff for rectangular matrix multiplication in VLSI models
\textit{J. E. Savage} [(*) J. Comput. Syst. Sci. 22, 230-242 (1981; Zbl 0458.68009)] has shown that the designer of any circuit for multiplying an \(m\times n\) matrix by an \(n\times p\) matrix ((m,n,p) problem) is confronted with an area-time tradeoff expressed by \(AT^ 2=\Omega(m^ 2p^ 2)\), when \((a-n)(b-n)<{1\over2}n^ 2\) with \(a=\max(m,n)\) and \(b=\max(n,p)\). No lower bound is known for the problem (m,n,p) when \((a- n)\times(b-n)\geq {1\over2}n^ 2\). In this paper the result given in (*) is partially emended and a new lower bound is given which holds for any choice of m,n,p, with \(m\geq n\), \(p\geq n\).
- Area-period tradeoffs for multiplication of rectangular matrices
- Area-time tradeoffs for bilinear forms computations in VLSI
- Area-time complexity of the unconstrained minimization problem
- Area—Time optimal VLSI integer multiplier with minimum computation time
- scientific article; zbMATH DE number 3890726 (Why is no real title available?)
- The matrix equation MX + XN = B in the VLSI model
This page was built for publication: Area-time tradeoff for rectangular matrix multiplication in VLSI models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q796300)
