Hi,
I'm doing some work with a genomics institute that is currently running single-threaded code built with gcc on a Dell cluster under Linux.
The work I'm going to do will involve building the code withthe Intel C++ compiler and parallelising some of the code using OMP.
Are there any recommendations of what version of Linuxto use with the most recent Intel C++ compiler on a Dell cluster?
I think theopenmp library supplied with the Intel compiler will look after the OMP interface. Are there any restrictions on the implementationof POSIX threads required to provide the underlying threads?
And do the performance tools (VTune, TBB etc) further affect the Linux/POSIX version required?
Here's hoping I don't get caught in a quagmire of software incompatibilities
Regards
David