Hello, it's now been a couple of times that I heard people saying that they heard of cases when TBB was used in hybrid distributed HPC applications, but they did not know the applications names. To me this is somewhat surprising (but like, a good surprise), because MPI applications are usually written in plain C or Fortran and if there is a computational loop to be parallelized using threads, programmers usually stick with OpenMP. However I can imagine that there are MPI applications written in C++ (otherwise why would there be C++ bindings in every MPI implementation for such a long time if no one used them) and they use C++ threading library, like TBB for intra-node communication.
So I was wondering if there is anyone here actually from HPC community or anyone heard of TBB being used as an intra-node threading framework for a real-world multi-process MPI application. I'd be so glad to know the names of these applications...
Robert, thanks, thats an interesting finding, I'll take a look at it. My googling for "tbb and mpi" returned no reasonable results, so I decided to ask on this forum. Probably searching for projects that use boost::mpi and then searching if those use TBB is a good idea, I know those at least will be C++ applications. I'll try that too.