Solved it myself.
According to Nvidia:
Starting with the Release 302 drivers, appl ication developers can direct the Optimus driver at runtime to use the High Performanc e Graphics to render any application–even those applications for which there is no existi ng application profile. They can do this by exporting a global variable named NvOptimusEnablement .
So applications can explicitly switch on my GPU Cores and use them.
I still don't know, what the connection between OpenMPI and my Nvidia Graphics card is, however I was able solving the problem by reinstalling a new Nvidia Graphics Card driver.