# Strong scaling on Summit **URL:** \ **Category:** General\ **Created:** [9 March 2020 14:09 UTC](https://pyfr.discourse.group/t/strong-scaling-on-summit/276 "2020-03-09T14:09:55Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Robert\_Sawko1](https://avatars.discourse-cdn.com/v4/letter/r/9de053/32.png) [@Robert\_Sawko1](https://pyfr.discourse.group/u/Robert_Sawko1)\ **Post date:** [9 March 2020 14:09 UTC](https://pyfr.discourse.group/t/strong-scaling-on-summit/276/1 "2020-03-09T14:09:55Z")
Dear PyFR developers, My colleague and I would like to setup two separate projects with PyFR: 1 external flow over a detailed car geometry, 2 internal incompressible flow through pipe bends with strong buoyancy. I have asked about the latter before, although never managed to run appropriate tests since then. We now have some preliminary tests on our medium size internal P9+6xV100 cluster and we see perfect scaling. Currently, we are evaluating whether our applications could benefit from Summit. My question to you is whether any Summit results have been published or presented in some format already? I would assume you have run benchmarks there. We want to avoid work repetition if possible and are interested if there are any lessons to be learnt for such installations. Please let us know. Best wishes, Robert ---
**Author:** ![fdw](https://avatars.discourse-cdn.com/v4/letter/f/a587f6/32.png) [@fdw](https://pyfr.discourse.group/u/fdw)\ **Post date:** [11 March 2020 14:49 UTC](https://pyfr.discourse.group/t/strong-scaling-on-summit/276/2 "2020-03-11T14:49:38Z")
Hi Robert, [pyfr-scaling-performance.pdf](https://pyfr.discourse.group/uploads/short-url/d95p2d0H9BErL9gY4kygZ80vIY7.pdf) (421 KB) ---
**Author:** ![Robert\_Sawko1](https://avatars.discourse-cdn.com/v4/letter/r/9de053/32.png) [@Robert\_Sawko1](https://pyfr.discourse.group/u/Robert_Sawko1)\ **Post date:** [12 March 2020 09:18 UTC](https://pyfr.discourse.group/t/strong-scaling-on-summit/276/3 "2020-03-12T09:18:37Z")
Freddie, Thanks for this. This is precisely what I was looking for. I've been also testing CUDA Aware MPI but on simpler benchmarks, but internally rather than on Summit. Best wishes, Robert