Optimizing using Intel Fortran Compiler 6.0

[philip-cobb7]

I posted this message on the Windows board because I didn't see this forum. I am actually running the Intel Fortran Compiler 6.0 for Linux and am very satisfied with it. I was wondering where I could get information on optimizing my code. I am running the code on two machines. One is an Intel Pentium 4 machine, and the other is a dual Intel Pentium 3 machine. What are some good optimization techniques to get my code running faster? Where can I get inforation about this? Thanks for the help.

[Steven_L_Intel1]

Welcome to the forum!

Have you studied the chapters on Optimization and Parallelization in the User's Guide? These discuss compiler switches and options of interest. For Pentium 4, the switches you'd want are "-tpp7 -xW" Note that the -xW generates code that requires a Pentium 4 to run. If you will be running this on other processors as well, use -axW instead - this will generate self-dispatching code.

The User's Manual will also discuss using OpenMP for parallelism on multiprocessor systems.

Steve

[philip-cobb7]

It would have been good if you had put a link to the FORTRAN compiler user guide instead of the C++ compiler, but I found it and am currently reading through the optimization section.