Someone has repealed Amdahl's law?
Once anything has to be shared the real ability to parallelize a problem goes right down.
Dr DK Panda is a world-recognised expert on parallel programming and networking. He's a Distinguished Scholar at The Ohio State University and his research group has developed the MVAPICH2 (high performance MPI and MIP+PGAS) libraries for InfiniBand, iWARP, and RoCE with support for GPUs, Xeon Phi, and virtualization. His …
The first step to solve a problem is to truly understand it. Once you have understood it, you might be able to reformulate it in such a way as to maximize the parallelization opportunities. If this does not happen then perhaps 1) the problem is inherently difficult to parallelize or 2) you are simply not good enough. I know which one it is for me, which is why I'm looking forward to watching this presentation.