Direct method solver based on multifrontal method

the high-speed and robust matrix solver

	Direct method solver based on multifrontal method
	Super Matrix Solver MF


	Special features of Super Matrix Solver MF

- Direct method based on multiforntal algorithm.
- Accelerated calculation speed by reduction of fill-in through algorithm (ordering) improvement.
- Easy-to-use with few parameters.
- Stably calculates a wide range of matrices such as those without diagonal dominant that cannot be solved with iterative method.


	MF Summary Specifications

Items	Descriptions	Notes
Target Analysis Fields	Structural Analyses and Electromagnetic, etc.
Target Coefficient Matrix	Sparse matrices that are generated from discretization methods such as finite element, finite volume and differential methods.	Calculates dense matrices; however, performs well for sparse matrices.
Problems with Zero Diagonal Elements	Capable of calculating.	Also calculates problems without diagonal dominant.
Types of Unknowns	Double precision real/complex number
Symmetry of Problems	Calculate both symmetric and asymmetric problems.	Also calculates hermitian and skew symmetric problems.
Calculable matrix scale	In the case of 1GB memory usage, approximate standard size is as follows: [2 dimension problems] * about 1,000,000 unknowns (in-core) * about 5,000,000 unknowns (out-of-core) [3 dimension problems] * about 200,000 unknowns (in-core) * about 800,000 unknowns (out-of-core)	The limit size can vary several times depending on complexity of particular grid.
Input Data	Coefficient matrix and right-hand-side vector
Output Data	Solution vector
Indication of Error Messages	Warnings and error messages returned as return value (calculation information, system information, etc).
Method of Provision	DLL format for Windows; Static library format for Linux and UNIX.	Source code will not be disclosed.
Attached Materials	Product manual (with explanations about data format, parameters, integration procedures, etc.), sample data, sample program for integrating SMS-AMG (C and FORTRAN).


	MF Performance Examples

Comparison of MF versus Other Solution Method

(Suppose the Super Matrix Solver MF's Fill-in = 1)

Sparse LU and Other MF Method were compared based on the fill-in performance, not on actual calculation time. Fill-in performance for Spares LU and Other MF method are published in paper, and therefore the comparison was made on the same fill-in performance. Theoretically, the fewer the number of fill-ins, the faster the calculation speed.


	More Information (OS)

- Super Matrix Solver-MF Product Info. (PDF/351KB)


	Platforms (OS)

Windows, Linux; for detailed information: MF System Environment (PDF/13KB)

NOTE: Materials provided on this web site does not guarantee functions and performance of the product.
Product specifications may change without notice.