# \#hpc **URL:** https://pyfr.discourse.group/tag/hpc/8.md [Latest](https://pyfr.discourse.group/latest.md) · [Categories](https://pyfr.discourse.group/categories.md) · [Tags](https://pyfr.discourse.group/tags.md) --- ## [Compute Q-criterion in Ascent and slice multiple times](https://pyfr.discourse.group/t/compute-q-criterion-in-ascent-and-slice-multiple-times/1337)
**Author:** [@AlmostSurelyRob](https://pyfr.discourse.group/u/AlmostSurelyRob)\ **Replies:** 2\ **Last updated:** [4 March 2026 16:57 UTC](https://pyfr.discourse.group/t/compute-q-criterion-in-ascent-and-slice-multiple-times/1337 "2026-03-04T16:57:15Z")
I would like to compute Q-criterion and visualise it through multiple slices. Doing this in Ascent would require me to create a pipeline and a scene for each slice. My current reading of the functionality is each pipeli… --- ## [PyFR plugin in-memory execution model](https://pyfr.discourse.group/t/pyfr-plugin-in-memory-execution-model/1335)
**Author:** [@AlmostSurelyRob](https://pyfr.discourse.group/u/AlmostSurelyRob)\ **Replies:** 1\ **Last updated:** [15 February 2026 17:29 UTC](https://pyfr.discourse.group/t/pyfr-plugin-in-memory-execution-model/1335 "2026-02-15T17:29:05Z")
I now have some semi-production runs and I am looking to optimise collection of statistics, integrals and point sampling information. I was warned that this will be costly in terms of performance, but I was very surprise… --- ## [Time step formatting is failing in writer plugin](https://pyfr.discourse.group/t/time-step-formatting-is-failing-in-writer-plugin/1317)
**Author:** [@AlmostSurelyRob](https://pyfr.discourse.group/u/AlmostSurelyRob)\ **Replies:** 6\ **Last updated:** [11 November 2025 17:15 UTC](https://pyfr.discourse.group/t/time-step-formatting-is-failing-in-writer-plugin/1317 "2025-11-11T17:15:09Z")
I am on develop branch of PyFR running a case with the following setting \[soln-plugin-writer\] dt-out = 10 basedir = . basename = k0-{t:06.1f} But the writer plugin is producing: ls -lt \*.pyfrs Nov 10 22:09 k0-0000.0.… --- ## [Environment for OLCF Frontier for develop branch?](https://pyfr.discourse.group/t/environment-for-olcf-frontier-for-develop-branch/1319)
**Author:** [@AlmostSurelyRob](https://pyfr.discourse.group/u/AlmostSurelyRob)\ **Replies:** 3\ **Last updated:** [11 November 2025 16:32 UTC](https://pyfr.discourse.group/t/environment-for-olcf-frontier-for-develop-branch/1319 "2025-11-11T16:32:58Z")
Hello, Could you advise on what would be a good environment setup for OLCF Frontier to run the develop branch of PyFR? I’ve configured everything I needed (see below) but when launching my test job with 8 rank/single no… --- ## [Scotch partitioning doesn't work out of the box](https://pyfr.discourse.group/t/scotch-partitioning-doesnt-work-out-of-the-box/1299)
**Author:** [@AlmostSurelyRob](https://pyfr.discourse.group/u/AlmostSurelyRob)\ **Replies:** 1\ **Last updated:** [18 September 2025 18:42 UTC](https://pyfr.discourse.group/t/scotch-partitioning-doesnt-work-out-of-the-box/1299 "2025-09-18T18:42:22Z")
I couldn’t immediately find an answer on the forum, so I thought writing this up may help another person. A similar error has been reported here but I thought a more explicit solution may be useful to some. Problem: Whe… --- ## [GPU parallelization and scalability](https://pyfr.discourse.group/t/gpu-parallelization-and-scalability/1243)
**Author:** [@Luca\_P](https://pyfr.discourse.group/u/Luca_P)\ **Replies:** 2\ **Last updated:** [7 April 2025 18:43 UTC](https://pyfr.discourse.group/t/gpu-parallelization-and-scalability/1243 "2025-04-07T18:43:00Z")
I was performing a simulation of the 2D flow around an airfoil on an HPC. I’m using a 3rd order type C mesh with a total number of nodes equal to 350964,also order=3 in the \[solver\] option and a dt=1e-7 with euler sche… --- ## [Partition error with scotch](https://pyfr.discourse.group/t/partition-error-with-scotch/1237)
**Author:** [@iagyonwu](https://pyfr.discourse.group/u/iagyonwu)\ **Replies:** 1\ **Last updated:** [25 March 2025 09:47 UTC](https://pyfr.discourse.group/t/partition-error-with-scotch/1237 "2025-03-25T09:47:15Z")
Hello, I am trying the provided tutorial on HPC (HLRS hunter: HLRS High Performance Computing Center Stuttgart: HPE Cray EX4000 (Hunter)). In a interactive session, I was able to tun the tutorial with one GPU with the… --- ## [Install pyfr on HPC with AMD gpu](https://pyfr.discourse.group/t/install-pyfr-on-hpc-with-amd-gpu/1234)
**Author:** [@iagyonwu](https://pyfr.discourse.group/u/iagyonwu)\ **Replies:** 2\ **Last updated:** [18 March 2025 13:32 UTC](https://pyfr.discourse.group/t/install-pyfr-on-hpc-with-amd-gpu/1234 "2025-03-18T13:32:07Z")
Hello everyone, I am new to PyFR, and tried to intall it on HPC (hlrs hunter: HPE Hunter Hardware and Architecture - HLRS Platforms). The HPC uses AMD mi300a processor. I followed the instruction on the Installation p… --- ## [Running PyFR on multiple GPUs on HPC cluster](https://pyfr.discourse.group/t/running-pyfr-on-multiple-gpus-on-hpc-cluster/1136)
**Author:** [@pyfr\_starting\_man](https://pyfr.discourse.group/u/pyfr_starting_man)\ **Replies:** 10\ **Last updated:** [18 July 2024 02:32 UTC](https://pyfr.discourse.group/t/running-pyfr-on-multiple-gpus-on-hpc-cluster/1136 "2024-07-18T02:32:18Z")
I have read through similar discussions on this, and I have seen to change device-id to local-rank for \[backend-cuda\] and to change GPUs to “compute exclusive mode”. However, I am running PyFR on supercomputer HPC cluste… --- ## [Scaling data on HPC clusters](https://pyfr.discourse.group/t/scaling-data-on-hpc-clusters/1104)
**Author:** [@Frankx9](https://pyfr.discourse.group/u/Frankx9)\ **Replies:** 3\ **Last updated:** [12 June 2024 14:24 UTC](https://pyfr.discourse.group/t/scaling-data-on-hpc-clusters/1104 "2024-06-12T14:24:34Z")
Hello, is there any scaling data for HPC clusters with PyFR which is relatively recent on AMD or NVIDIA vendors? Best --- ## [PyFR 1.15.0 Run Issue: libxsmm](https://pyfr.discourse.group/t/pyfr-1-15-0-run-issue-libxsmm/853)
**Author:** [@Merc96](https://pyfr.discourse.group/u/Merc96)\ **Replies:** 6\ **Last updated:** [11 March 2024 17:35 UTC](https://pyfr.discourse.group/t/pyfr-1-15-0-run-issue-libxsmm/853 "2024-03-11T17:35:47Z")
Hi All, I’m trying run Pyfr on Rhel8.5. The other packages versions: Python 3.9.16 libxsmm-1.17 gcc-8.5.0 (RHEL8.5 Default) OpenMPI-4.1.2 PyFR 1.15.0 (install ed with Pip) I built the libxsmm from source with docume… --- ## [How to speed up a compressible T106C simulation](https://pyfr.discourse.group/t/how-to-speed-up-a-compressible-t106c-simulation/776)
**Author:** [@ManuSaiVenkataRatnam](https://pyfr.discourse.group/u/ManuSaiVenkataRatnam)\ **Replies:** 5\ **Last updated:** [27 October 2023 04:08 UTC](https://pyfr.discourse.group/t/how-to-speed-up-a-compressible-t106c-simulation/776 "2023-10-27T04:08:51Z")
Hello Team Pyfr, This is in regards to my thesis work. I have to do T106C test case for my work, but the simulation is taking more time for a warm up of 100 time units of order 1. following are some initial test runs… --- ## [pyFR encounters a blocking issue when using Slurm](https://pyfr.discourse.group/t/pyfr-encounters-a-blocking-issue-when-using-slurm/952)
**Author:** [@studyhardhardhard](https://pyfr.discourse.group/u/studyhardhardhard)\ **Replies:** 2\ **Last updated:** [25 October 2023 17:19 UTC](https://pyfr.discourse.group/t/pyfr-encounters-a-blocking-issue-when-using-slurm/952 "2023-10-25T17:19:57Z")
Hello, everyone. I have been using Slurm to configure multiple GPU resources for running pyFR simulations recently. Through multi-GPU coordination, pyFR has been performing simulation tasks successfully and yielding sat… --- ## [Rough estimate for cost of DNS at Re=400000](https://pyfr.discourse.group/t/rough-estimate-for-cost-of-dns-at-re-400000/883)
**Author:** [@steve](https://pyfr.discourse.group/u/steve)\ **Replies:** 2\ **Last updated:** [25 July 2023 16:22 UTC](https://pyfr.discourse.group/t/rough-estimate-for-cost-of-dns-at-re-400000/883 "2023-07-25T16:22:28Z")
I am looking for an order of magnitude estimate for the cost or running a scale resolved DNS solution on an object with transition and significant separation at a Re of 400k. For units I am thinking perhaps A100 hours m… --- ## [Is using partitioning combined with CUDA the fastest computing method?](https://pyfr.discourse.group/t/is-using-partitioning-combined-with-cuda-the-fastest-computing-method/864)
**Author:** [@studyhardhardhard](https://pyfr.discourse.group/u/studyhardhardhard)\ **Replies:** 5\ **Last updated:** [10 July 2023 16:54 UTC](https://pyfr.discourse.group/t/is-using-partitioning-combined-with-cuda-the-fastest-computing-method/864 "2023-07-10T16:54:29Z")
Hello everyone. I am a newcomer to pyFR, but I want to grasp it quickly. I’m attempting to run a three-dimensional case based on the paper “On the utility of GPU accelerated high-order methods forunsteady flow simulati… --- ## [MPI collect wait times informations](https://pyfr.discourse.group/t/mpi-collect-wait-times-informations/838)
**Author:** [@Frankx9](https://pyfr.discourse.group/u/Frankx9)\ **Replies:** 24\ **Last updated:** [14 June 2023 14:38 UTC](https://pyfr.discourse.group/t/mpi-collect-wait-times-informations/838 "2023-06-14T14:38:51Z")
Hi, I was trying to detect load imbalances but when using collect wait times in the backed section, my solution file only has these informations: \[data\] fields = rho,rhou,rhov,rhow,E prefix = soln \[backend\] rank-alloc… --- ## [Poor TGV performance with GPU and MPI](https://pyfr.discourse.group/t/poor-tgv-performance-with-gpu-and-mpi/820)
**Author:** [@luli](https://pyfr.discourse.group/u/luli)\ **Replies:** 5\ **Last updated:** [1 June 2023 13:56 UTC](https://pyfr.discourse.group/t/poor-tgv-performance-with-gpu-and-mpi/820 "2023-06-01T13:56:31Z")
When I run the case tgv in TGV Performance Numbers - General - PyFR with config file as follow: \[backend\] precision =double rank-allocator = linear \[backend-cuda\] device-id = local-rank \[constants\] gamma = 1.4 mu = 6.… --- ## [Multi GPU calculation](https://pyfr.discourse.group/t/multi-gpu-calculation/792)
**Author:** [@Yujif1Aero](https://pyfr.discourse.group/u/Yujif1Aero)\ **Replies:** 5\ **Last updated:** [10 April 2023 04:56 UTC](https://pyfr.discourse.group/t/multi-gpu-calculation/792 "2023-04-10T04:56:47Z")
Dear All, I have a question about multi-GPU calculation. My understading is that PyFR is able to calculate CFD with 2 Nvidia GPUs when we put “$mpiexec -n 2 pyfr run -b cuda -p” on Bash. But my task manager shows "G… --- ## [How to set openmpi](https://pyfr.discourse.group/t/how-to-set-openmpi/777)
**Author:** [@fei](https://pyfr.discourse.group/u/fei)\ **Replies:** 1\ **Last updated:** [18 March 2023 14:07 UTC](https://pyfr.discourse.group/t/how-to-set-openmpi/777 "2023-03-18T14:07:38Z")
Hello, First computer with one socket and 12 cores, I run with “pyfr run -b openmp -p \*.pyfrm \*.ini”, the calculation time is approximately 130000. Second computer with two sockets and 32 cores, as the performance tuni… --- ## [CPU parallelization problem for tutorial euler case](https://pyfr.discourse.group/t/cpu-parallelization-problem-for-tutorial-euler-case/470)
**Author:** [@verseau](https://pyfr.discourse.group/u/verseau)\ **Replies:** 4\ **Last updated:** [27 August 2021 11:26 UTC](https://pyfr.discourse.group/t/cpu-parallelization-problem-for-tutorial-euler-case/470 "2021-08-27T11:26:36Z")
Hi, I am running the euler\_vortex\_2d case. I used the original mesh and .ini file. It seemed to work fine with cuda: mpiexec -n 2 pyfr run -b cuda -p euler\_vortex\_2d.pyfrm euler\_vortex\_2d.ini took about 1 minute mpie… --- ## [Scalability and linearised NS solver](https://pyfr.discourse.group/t/scalability-and-linearised-ns-solver/464)
**Author:** [@Zhenyang](https://pyfr.discourse.group/u/Zhenyang)\ **Replies:** 5\ **Last updated:** [18 August 2021 15:56 UTC](https://pyfr.discourse.group/t/scalability-and-linearised-ns-solver/464 "2021-08-18T15:56:41Z")
Hi, I am interested in the optimal grid points for each CPU/GPU core since I am going to run some large scale cases. Although it could be different from machines, it would be nice to have some basic idea before doing te…