Matrix solver for Impicit scheme

harish_0689@ · December 27, 2011, 08:47

Hello, I am developing an implicit reimann solver for Euler equations and i am using GMRES algorithm to solve the matrix, but i find that Implicit is taking more CPU time than explicit solver and may be the problem is with the matrix solver. I need some suggestions regarding it as which is the best solver for the implicit schemes and sources from which i can get it?

duri · December 27, 2011, 10:17

Take advantage of sparse system. Implement GMRES for sparse system.
GMRES can even solve indefinite system. This much advanced solver in not required for Euler equations. Lot of simple, robust solvers are available and they are fast enough. I recommend sparse PCG solver.

harish_0689@ · December 28, 2011, 08:06

@duri: thanx a lot for the reply...My guide actually suggested me to try out Gauss-seidel point relaxation method..but d problem is i am not able to find the readymade solver..can you suggest me some source from which i can get these readymade PCG or gauss-seidel point relaxation solvers n possibly in C language.

ganesh · December 28, 2011, 09:17

Dear Harish,

A good open-source package with a large number of built-in linear solvers is LiS. You can find it with documentation at www.ssisc.org/lis. It supports both F90 and C programming languages, but there is not much user-support available. The manual however is self-explanatory and should do the job.

Regards,

Ganesh

harish_0689@ · December 28, 2011, 17:56

Thank u very much ganesh. I found it and seems to be helpful..i will try it out.

harish_0689@ · January 5, 2012, 03:46

@ganesh: I an not able to understand how to integrate the linear solvers you mentioned into my code. Can u please just brief me how to do so..

December 27, 2011, 08:47	Matrix solver for Impicit scheme	#1
harish_0689@ New Member C.Harish Join Date: Jun 2011 Posts: 15 Rep Power: 14	Hello, I am developing an implicit reimann solver for Euler equations and i am using GMRES algorithm to solve the matrix, but i find that Implicit is taking more CPU time than explicit solver and may be the problem is with the matrix solver. I need some suggestions regarding it as which is the best solver for the implicit schemes and sources from which i can get it?

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
different results between serial solver and parallel solver	wlt_1985	FLUENT	11	October 12, 2018 08:23
2nd order upwind scheme (Fluent and CFX)	Far	FLUENT	0	May 22, 2011 01:50
C++ solver for Large sparse matrix!	sina_mech	Main CFD Forum	7	November 20, 2009 13:52
Solver scheme	Richard	CFX	2	March 9, 2006 08:55
solver for linear system with large sparse matrix	Yangang Bao	Main CFD Forum	1	October 25, 1999 04:22

December 27, 2011, 10:17		#2
duri Senior Member duri Join Date: May 2010 Posts: 245 Rep Power: 16	Take advantage of sparse system. Implement GMRES for sparse system. GMRES can even solve indefinite system. This much advanced solver in not required for Euler equations. Lot of simple, robust solvers are available and they are fast enough. I recommend sparse PCG solver.

December 28, 2011, 08:06		#3
harish_0689@ New Member C.Harish Join Date: Jun 2011 Posts: 15 Rep Power: 14	@duri: thanx a lot for the reply...My guide actually suggested me to try out Gauss-seidel point relaxation method..but d problem is i am not able to find the readymade solver..can you suggest me some source from which i can get these readymade PCG or gauss-seidel point relaxation solvers n possibly in C language.

December 28, 2011, 09:17		#4
ganesh Member ganesh Join Date: Mar 2009 Posts: 40 Rep Power: 17	Dear Harish, A good open-source package with a large number of built-in linear solvers is LiS. You can find it with documentation at www.ssisc.org/lis. It supports both F90 and C programming languages, but there is not much user-support available. The manual however is self-explanatory and should do the job. Regards, Ganesh

December 28, 2011, 17:56		#5
harish_0689@ New Member C.Harish Join Date: Jun 2011 Posts: 15 Rep Power: 14	Thank u very much ganesh. I found it and seems to be helpful..i will try it out.

January 5, 2012, 03:46		#6
harish_0689@ New Member C.Harish Join Date: Jun 2011 Posts: 15 Rep Power: 14	@ganesh: I an not able to understand how to integrate the linear solvers you mentioned into my code. Can u please just brief me how to do so..