Open Access
Table 3
Configuration of the clusters used for the performance tests, and compilation options used to compile IDEFIX.
Cluster | Node configuration | Compiler/libraries | Compiler option |
---|---|---|---|
Irene Rome TGCC | 2×AMD EPYC 7H12 (128 cores at 2.6 GHz) | Intel 2020.4 OpenMPI 4.1.4 | -O3 -avx2 -std=gnu++17 |
Jean Zay CSL IDRIS | 2×Intel Cascade Lake 6248 (40 cores at 2.5 GHz) | Intel 2020.4 Intel MPI 2019.9 | -O3 -xCORE-AVX512 -std=gnu++17 |
Jean Zay V100 IDRIS | 4×NVidia Tesla V100 +2×Intel Cascade Lake 6248 | nvcc/CUDA 11.2 OpenMPI 4.1.1 | -O3 -expt-extended-lambda -arch=sm_70 -std=c++17 |
AdAstra CINES | 4×AMD Mi250X +2×AMD EPYC 7713 | Cray PE 2.7.19 Cray Mpich 8.1.21 Rocm 5.2.0 | -O3 -fno-gpu-rdc -x hip -offload-arch=gfx90a -std=gnu++17 |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.