Dear Open MPI developers, I'm a graduate student and recently test some minibenchmarks using Open MPI library on Xeon X5650 cluster. From the website of Open MPI, I know that Open MPI can use
*sm BTL* to transfer data with high bandwidth via shared memory. My question is that what about Open MPI collectives on shared memory? Were they implemented and optimized on top of point-to-point communication or utilizing shared memory separately? Best Regards, Shigang Li.