Posted: 5 Jul 2013 | 16:06
Have you ever wanted to send a message using MPI to a specific thread in a multi-threaded MPI process? With the current MPI Standard, there is no way to identify one thread from another. The whole MPI process has a single rank in each communicator.
Posted: 3 Jul 2013 | 08:38
This post was written by Jon Hill from Imperial College, who used to work at EPCC and has been collaborating with us recently.
I always jump at the chance to work with EPCC. Not just because they are my former employers (ah, the joys of Friday buns). Nor is it due to Edinburgh being one of my favourite cities and collaborating with EPCC is a good excuse to visit. The main reason for collaborating with EPCC is to use the wealth of experience the people working there have on making scientific code go even faster. Whilst this is extremely important to our research, we don't have the time to do both science and improve code performance.
Posted: 2 Jul 2013 | 16:12
Twenty four students from all over Europe are just embarking on the PRACE Summer of HPC. Before splitting up and heading off to 10 different European HPC centres to spend two months working on their HPC projects, they are all gathered together here at EPCC to do an intensive week's HPC training as part of the PRACE Advanced Training Centre programme.
Here they are resplendent in their SoHPC T-shirts ...