commons-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Haswell, Joe" <>
Subject RE: [MATH] Linear equation [A] [X] = [B] solution using Sparse Matrix
Date Wed, 08 Dec 2010 21:50:51 GMT
How large?  It's likely one of those pesky time-complexity things you run into sometimes =(

Joe H.  | HP Software

-----Original Message-----
From: Mike Zhou [] 
Sent: Wednesday, December 08, 2010 2:49 PM
To: Commons Users List
Subject: [MATH] Linear equation [A] [X] = [B] solution using Sparse Matrix


I am trying to use common math library to solve large-scale liner equation
with sparse structure. Use the SpareRealMatrix interface
and OpenMapRealMatrix implementation, I defined my matrix. Then using the LU

new LUDecompositionImpl(A).getSolver().solve(B)

I could find the solution. However, the LU based solution seems to be very
slow. I guess, it solves the sparse matrix as a full matrix.


- Does the LUComposition implementation take full advantage of the sparse
matrix structure
- If yes, where I might be wrong?

I am using common math lib 2.1.

Thanks, Mike Zhou

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message