Equation Solution  
    High Performance by Design
Navigation Tree  
Home  
'- - Programming Tools
'- - Parallel Computing
|     '- - LAIPE
|     '- - LAIPE2
|     |     '- - System Equations
|     |     '- - Auxiliary Routines
|     |     '- - Matrix Product
|     '- - neuLoop
|     '- - MTASK
'- - Blog
'- - In-Situ Evaluation
'- - Numerical Analysis
'- - Structural Mechanics
|     '- - IFAS
|     '- - mCable
'- - Write Us
|     '- - Feedback  
|     '- - Info  
|     '- - Support  
|     '- - Webmaster  
'- - Privacy Policy



LAIPE2: parallel package for numerical analysis



      LAIPE2, is the second-generation LAIPE, which stands for Link And In Parallel Execute. The original LAIPE is programmed in MTASK, multitasking technology; LAIPE2 is an application of neuLoop, soft core computing. The LAIPE2 is included in the gfortran distributed on this site.

      When linking an application against LAIPE2, it is also necessary to link against neuLoop. For example,

   gfortran your_program -llaipe2 -lneuloop4
 or
   gfortran your_application -llaipe2 -lneuloop6

where neuloop4 is a library for homogenous computing and neuloop6 is for heterogeneous computing. The libraries, neuloop4 and neuloop6, are also included in gfortran for free use.

Known Issues
  • The 32-bit version has bugs with COMPLEX*20 computing (e.g, extended precision COMPLEX variables). The bug is originated in gfortran. Single precision, double precision and quad precision do not have the bug.
  • The 64-bit version does not have bug.