I make the numerical calculation program and run two computers. (One is used i7 cpu, others Xeon cpu.)
The numerical result is slightly different. (input data is same).
My program is very sensitive to numerical difference, so i confused which one is correct.
Why the differece is occured?
the program is complied same computer under environment
Complier : MS VS2010 C++ Complier
MKL library Version : 184.108.40.206
Linked library : mkl_core.lib, mkl_intel_lp64.lib, mkl_intel_thread.lib, libiomp5md.lib
the program mainly used "DGEMM", "PARDISO" functions for solving linear equation directly(not iterative method).
And the number of thread set 1.
yes, this is expected behavior with floating point computations. since version 11.1, MKL addresses such issues. Please refer to the documentation about CNR - https://software.intel.com/en-us/node/528577