Title
An Approximate Minimum Degree Ordering Algorithm
Abstract
An approximate minimum degree (AMD) ordering algorithm for preordering a symmetric sparse matrix prior to numerical factorization is presented. We use techniques based on the quotient graph for matrix factorization that allow us to obtain computationally cheap bounds for the minimum degree. We show that these bounds are often equal to the actual degree. The resulting algorithm is typically much faster than previous minimum degree ordering algorithms and produces results that are comparable in quality with the best orderings from other minimum degree algorithms.
Year
DOI
Venue
1996
10.1137/S0895479894278952
SIAM J. Matrix Analysis Applications
Keywords
Field
DocType
previous minimum degree,symmetric sparse matrix,numerical factorization,approximate minimum degree,resulting algorithm,matrix factorization,minimum degree algorithm,approximate minimum degree ordering,actual degree,minimum degree,best ordering,sparse matrix,sparse matrices
Discrete mathematics,Mathematical optimization,Matrix decomposition,Minimum degree algorithm,Algorithm,Cuthill–McKee algorithm,Degree (graph theory),Degree matrix,Factorization,Quotient graph,Sparse matrix,Mathematics
Journal
Volume
Issue
ISSN
17
4
0895-4798
Citations 
PageRank 
References 
254
22.83
7
Authors
3
Search Limit
100254
Name
Order
Citations
PageRank
Patrick R. Amestoy144644.24
TIMOTHY A. DAVIS21447144.19
Iain S. Duff31107148.90