The following is a compilation of the Intel® MKL PARDISO related articles from the Knowledge Base.
- Tips for using PARDISO : Provides tips for getting started and corrects some of the common mistakes made by first-time and experienced users.
- PARDISO Parameter Tables: PARDISO has many parameters that must be understood and properly set to get the expected results.
- Find Solution to Input Error in PARDISO* : Address a kind of common error like *** Error in PARDISO when call PARDISO* and provide common check points guide to solve the problem
- Description of PARDISO errors and messages: Detailed description of errors and messages returned by PARDISO.
- How to use OOC PARDISO? : How to enable OOC (out-of-core) version of PARDISO from Intel® MKL
- Parallel Direct Sparse Solver for Clusters: Introduced Cluster version of PARDISO in MKL 11.2
- Summary of the API differences between University of Basel (UB) PARDISO* and Intel® MKL PARDISO : Explains the differences between University of Basel PARDISO and Intel® MKL PARDISO.
- New ILP64 (64-bit integer) pardiso_64 interface of PARDISO is now available: ILP64 (64-bit integer ) version of the PARDISO, namely PARDISO_64, is has been added.
- PARDISO: incorrect behavior in deciding whether to use in-Core or out-of-core (OOC): One of the known issues in MKL 10.x and was fixed in MKL 10.2.5 and 10.3
- True DSS and PARDISO overloaded F90 interfaces are available in Intel MKL 10.3.1 and 10.2.7: Renewed F90 interfaces availability for PARDISO in MKL 10.3.1 and 10.2.7 onwards.
- PARDISO OOC supports the extended set of matrices which can be solved on IA32
- PARDISO new feature ( store and load handle on HDD ) – open discussion
- Dynamic MKL OOC PARDISO cannot solve large problem on Win64 bit OS.
- Intel® MKL Sparse Solvers Training Material