- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I am trying to use CXML's sparse matrix solvers. From the documentation
it is not clear if CXML has complex sparse iterative solvers. All the examples show 'real' sparse matrices. Does CXML has complex
sparse iterative solvers ?? If so any pointers in the documentation..
I could successfully use the complex direct sparse solvers in CXML.
But I want to use them as ILU preconditioners in my own sparse
iterative solvers. Is there a way I can set 'no extra fil in' option so
I have control over the memory usage.
I appreciate any help in this regard.
Thanks in advance.
C.J. Reddy
it is not clear if CXML has complex sparse iterative solvers. All the examples show 'real' sparse matrices. Does CXML has complex
sparse iterative solvers ?? If so any pointers in the documentation..
I could successfully use the complex direct sparse solvers in CXML.
But I want to use them as ILU preconditioners in my own sparse
iterative solvers. Is there a way I can set 'no extra fil in' option so
I have control over the memory usage.
I appreciate any help in this regard.
Thanks in advance.
C.J. Reddy
Link Copied
2 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
You also sent this to fortran@compaq.com, so consider this a reply from there too.
I asked the CXML lead engineer your questions. The answer to the first question is "No". He was baffled by your second question, saying, "Direct sparse solvers are designed to avoid creating "fill-in," so the idea of a way to specify not doing something they are designed not to do in the first place doesn't make sense."
Steve
I asked the CXML lead engineer your questions. The answer to the first question is "No". He was baffled by your second question, saying, "Direct sparse solvers are designed to avoid creating "fill-in," so the idea of a way to specify not doing something they are designed not to do in the first place doesn't make sense."
Steve
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks for the answers.
For the second question: Normally direct solvers are designed to 'minimize' extra
fill in. But in practice they do have some extra fill-in. My question was to see
if we can set this "fill-in" to zero (ofcourse the solution will not be right !), but
I can use this solution in my iterative solver as a preconditioner (normally referred
to as ILU preconditioner).
Thanks
C.J. Reddy
For the second question: Normally direct solvers are designed to 'minimize' extra
fill in. But in practice they do have some extra fill-in. My question was to see
if we can set this "fill-in" to zero (ofcourse the solution will not be right !), but
I can use this solution in my iterative solver as a preconditioner (normally referred
to as ILU preconditioner).
Thanks
C.J. Reddy
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page