Solves unsymmetric, structurally symmetric or symmetric systems, real or complex, positive definite or indefinite, hermitian.
LU decomposition with complete pivoting.
Parallel on SMPs and Cluster of SMPs.
Automatic combination of iterative and direct solver algorithms.
The PARDISO Version 5.0.0 has been released in January 2014. It contains full support of multi-threaded Schur-complement computations and full support for parallel selected inversion. The new features are described in the following papers:
- M. Luisier, O. Schenk et.al., Fast Methods for Computing Selected Elements of the Green's Function in Massively Parallel Nanoelectronic Device Simulations, Euro-Par 2013, LNCS 8097, F. Wolf, B. Mohr, and D. an Ney (Eds.), Springer-Verlag Berlin Heidelberg, pp. 533–544, 2013,
- C. Petra, O. Schenk, M. Lubin, K. Gärtner, An augmented incomplete factorization approach for computing the Schur complement in stochastic optimization, SIAM J. Sci. Comput, accepted, in press.
Additional publications related to PARDISO are available on the webpage of the research group of Olaf Schenk.
Important: Please note that the Intel MKL version of PARDISO is based on our version from 2006 and that a lot of new features and improvements of PARDISO are not available in the Intel MKL library.
How To Use PARDISO
- symmetric linear systems
- unsymmetric linear systems
- complex unsymmetric linear systems
- symmetric linear systems showing an example based on a discretization of the 2D Laplacian equation (laplace.c, laplace.h)
- scalable MPI C example for symmetric linear systems showing a parallel MPI example based on a discretization of the 2D Laplacian equation (laplace_mpi.tgz)
C++, Schur Complement
Julia interface for all types of linear systems in PARDISO
(by Kristoffer Carlsson , Chalmers University of Technology, Göteborg).
Current Available Libraries
Version 5.0.0 (Architecture x86-64, 64-bit)
|Compiler||Operating System||PARDISO Libraries|
|mpicc/mpif77 4.5.0||Linux, MPI||libpardiso500-MPI-GNU450-X86-64.so|
|mpicc/mpif77 4.6.1||Linux, MPI||libpardiso500-MPI-GNU461-X86-64.so|
|mpicc/mpif77 4.6.3||Linux, MPI||libpardiso500-MPI-GNU463-X86-64.so|
|mpicc/mpif77 4.7.2||Linux, MPI||libpardiso500-MPI-GNU472-X86-64.so|
|icc/ifort 13.0.1||Linux, MPI||libpardiso500-MPI-INTEL1301-X86-64.so|
(with optimized BLAS/LAPACK from Intel MKL)
|gcc/gfortran 4.7.1||MAC OSX 10.6.5||libpardiso500-MACOS-X86-64.dylib,
Other libraries can be compiled upon request.
Please let us know in case that some of these libraries are not working for you.
Ipopt 3.12.1 (extended version), PARDISO 5.1.0 (threaded version)
You need to set the numbers of threads with, for example, if you would like to use 8 cores you should use
export OMP_NUM_THREADS=8 before running ipopt.mexa64 in Matlab.
In addition, you need a PARDISO license file pardiso.lic which can be obtained here.
|Architecture x86-64, 64-bit, Matlab R2014b||Linux||ipopt.mexa64, ipopt_auxdata.m, ipopt.opt|
|Architecture x86-64, 64-bit, Matlab R2012b||Linux||ipopt.mexa64, ipopt_auxdata.m, ipopt.opt|
|Architecture x86-64, 64-bit, Matlab R2014b||Mac OS X 10.9.5||ipopt.mexmaci64, ipopt_auxdata.m, ipopt.opt|
Here you can dowload an academic license for the current release of PARDISO.
Please choose the desired license type:
Please contact email@example.com if you are interested in an evaluation license or other licenses for a company.
Release of new webpage, including user map and list of successful recent consulting projects.).
Release of Version 5.0.0, including new Schur-complement solver and selected inversion software (release notes).
Release of Version 4.1.2, including the distributed memory solver (release notes).
Release of Version 4.1.0 (release notes).
Map of pardiso users
Release of Version 4.0.0 (release notes).
The solver is now available for both academic and commercial use.
The solver PARDISO is now available under www.pardiso-project.org.
In case that you are using the new version 5.0.0 please cite:
M. Luisier, O. Schenk et.al.,
Fast Methods for Computing Selected Elements of the Green's Function in Massively Parallel Nanoelectronic Device Simulations,
Euro-Par 2013, LNCS 8097, F. Wolf, B. Mohr, and D. an Ney (Eds.), Springer-Verlag Berlin Heidelberg, pp. 533–544, 2013.
O. Schenk, M. Bollhöfer, and R. Römer,
On large-scale diagonalization techniques for the Anderson model of localization.
Featured SIGEST paper in the SIAM Review selected "on the basis of its exceptional interest to the entire SIAM community". SIAM Review 50 (2008), pp. 91-112.
O. Schenk, A. Wächter, and M. Hagemann,
Matching-based Preprocessing Algorithms to the Solution of Saddle-Point Problems in Large-Scale Nonconvex Interior-Point Optimization.
Journal of Computational Optimization and Applications, pp. 321-341, Volume 36, Numbers 2-3 / April, 2007.
For support and consulting questions related to PARDISO 5.0.0, please send an e-mail to
We provide consulting services in the field of computational mathematics, scientific computing and high-performance computing including:
Benchmarking services and optimal use of the solver software
Effective use of high performance computational systems
Re-optimization of successfully implemented systems
Completed projects and a few selected consulting examples (since 2013)
Training and general solver recommendation for precision measuring technology product simulations.
Marathon Oil Company
Training and general solver recommendation for the Marathon Oil Reservoir Simulation product.
Personnel training in the area of OpenCL and sparse solvers.
Integration of PARDISO 5.0.0 into CST's electromagnetic design software.
Integration of PARDISO 5.0.0 into MOSEK's optimization software for solving mathematical optimization problems.
AutoForm Engineering AG
Integration of PARDISO 5.0.0 into AutoForm's automobile sheet metal forming tools.
Endress+Hauser Flowtec AG, Reinach, Switzerland
Integration of PARDISO 5.0.0 into finite-element tools at Endress+Hauser Flowtec AG.
U.S. Naval Observatory
Integration of PARDISO 5.0.0 into simulation tools at the U.S. Naval Observatory.
SMS Siemag AG
Integration of PARDISO 5.0.0 into simulation tools for metallurgical plant and rolling mill technology.
Integration of PARDISO 5.0.0 into Frustum's Distributed Optimization Service.
Integration of PARDISO 5.0.0 into Silvaco simulation tools.
Integration of PARDISO 5.0.0 into JCMsuite, a complete Finite Element software package for the computation of electromagnetic waves, elasticity and heat conduction.
Integration of PARDISO 4.1.2 into reservoir simulation tools at Saudi Aramco.
Map of PARDISO Users (since 2014)
Authors & Contact
Institute of Computational Science, USI Lugano, Switzerland
WIAS Berlin, Germany
University of Minnessota, contributed with the METIS package
Technical University of Berlin, contributed with a matrix input checking tool
Dept. of Human Genetics, University of Chicago, contributed with an excellent MATLAB interface
Chalmers University of Technology, Göteborg, contributed with a Julia PARDISO interface