You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/software/sciapps/gromacs.md
+35-1Lines changed: 35 additions & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -159,9 +159,43 @@ Protein atoms = 86,996 Lipid atoms = 867,784 Water atoms = 2,041,230 Ions = 1
159
159
- If the problem allows the integration step to take place on the GPU with `-update gpu`, that can lead to significant performance and scaling gains as it allows an even greater part of the computations to take place on the GPU.
160
160
- A single node of the GH200 cluster offers 4x CPU+GPU. For problems that can benefit from scaling beyond a single node, use the flag `export FI_CXI_RX_MATCH_MODE=software` in the SBATCH script. The best use of resources in terms of node-hours might be achieved on a single node for most simulations.
161
161
162
+
### Building GROMACS from Source
163
+
164
+
The [GROMACS] uenv provides all the dependencies required to build GROMACS from source, with several optional features enabled. This approach is particularly useful for deploying customized variants of GROMACS. You can follow these steps to build GROMACS from source:
165
+
166
+
```bash
167
+
uenv start --view=develop <GROMACS_UENV># (1)!
168
+
169
+
cd<PATH_TO_GROMACS_SOURCE># (2)!
170
+
171
+
mkdir build &&cd build
172
+
cmake \
173
+
-DCMAKE_C_COMPILER=gcc \
174
+
-DCMAKE_CXX_COMPILER=g++ \
175
+
-DGMX_MPI=on \
176
+
-DGMX_GPU=CUDA \
177
+
-GMX_CUDA_TARGET_SM="90"\ # for the Hopper GPUs
178
+
-DGMX_SIMD=ARM_NEON_ASIMD \ # for the Grace CPUs
179
+
-DGMX_DOUBLE=off \ # turn on double precision only if useful
0 commit comments