메뉴
VASP on MIT Engaging. VASP 6.4.2 (GCC/OpenMPI, Rocky 8 compatible), module loading, SLURM submission, POTCAR paths. Use when setting up or running VASP calculations on Engaging.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | vasp-engaging |
| description | VASP on MIT Engaging. VASP 6.4.2 (GCC/OpenMPI, Rocky 8 compatible), module loading, SLURM submission, POTCAR paths. Use when setting up or running VASP calculations on Engaging. |
Compiled with GCC 12.2.0 + OpenMPI 4.1.4 on AMD EPYC (Rocky 8). Works on all current partitions without CentOS 7 constraints.
Binary: /nfs/rafagblab001/vasp.6.x.x/vasp.6.4.2/bin/vasp_std
Module: /nfs/rafagblab001/software/modulefiles/vasp/6.4.2
module use -a /nfs/rafagblab001/software/modulefiles
module load vasp/6.4.2
This automatically loads GCC 12.2.0, OpenMPI 4.1.4, and sets LD_LIBRARY_PATH for OpenBLAS, FFTW, and ScaLAPACK.
#!/bin/bash
#SBATCH -p mit_normal # or ou_cheme, mit_preemptable, pi_ccoley, etc.
#SBATCH --nodes=1
#SBATCH -n 4
#SBATCH --cpus-per-task=2
#SBATCH --mem-per-cpu=4G
#SBATCH -t 0-06:00:00
#SBATCH -J vasp_job
#SBATCH --output=%j_%x.log
export OMP_NUM_THREADS=$SLURM_CPUS_PER_TASK
module use -a /nfs/rafagblab001/software/modulefiles
module load vasp/6.4.2
cd $SLURM_SUBMIT_DIR
srun vasp_std
| Binary | Use Case |
|---|---|
vasp_std | Standard: general k-point meshes |
vasp_gam | Gamma-only: molecules/clusters in box (~2x faster at gamma) |
vasp_ncl | Non-collinear: SOC calculations |
POTCARs (PBE, v54) are at /nfs/rafagblab001/vasp_PBE_54/. Access requires membership in the sched_mit_rafagb group.
cat /nfs/rafagblab001/vasp_PBE_54/Ag/POTCAR \
/nfs/rafagblab001/vasp_PBE_54/O/POTCAR \
> POTCAR
Species order in POTCAR must match species order in POSCAR.
If you need to rebuild (e.g., for a new version), follow https://orcd-docs.mit.edu/recipes/build-vasp-gcc-cpu/:
# get an interactive session on an AMD EPYC node (where jobs will run)
salloc -p pi_ccoley -w node2519 -n 8 --mem=16G --time=1:00:00
module load gcc/12.2.0 openmpi/4.1.4 netlib-lapack/3.10.1 netlib-scalapack/2.2.0 fftw/3.3.10 openblas/0.3.26
cd /nfs/rafagblab001/vasp.6.x.x/vasp.6.4.2
cp arch/makefile.include.gnu_omp makefile.include
SCALAPACK_ROOT=$(module -t show netlib-scalapack 2>&1 | grep CMAKE_PREFIX_PATH | awk -F, '{print $2}' | awk -F\" '{print $2}')
FFTW_ROOT=$(pkgconf --variable=prefix fftw3)
OPENBLAS_ROOT=$(dirname $(pkgconf --variable=libdir openblas))
make -j 8 OPENBLAS_ROOT=$OPENBLAS_ROOT FFTW_ROOT=$FFTW_ROOT SCALAPACK_ROOT=$SCALAPACK_ROOT MODS=1 DEPS=1
Important: Always compile on the same CPU architecture where jobs will run. -march=native in the makefile bakes in the compile node's instruction set.
scp works but rsync may hang due to Duo 2FA. For directories, create the remote dir first:
ssh engaging1 "mkdir -p ~/path/to/remote/dir"
scp local/files/* engaging1:~/path/to/remote/dir/
The old VASP 6.2.1 at /nfs/rafagblab001/software/vasp/vasp.6.2.1/ is compiled against Intel MKL/MPI 2017 and only works on CentOS 7 nodes (sched_mit_rafagb with -C centos7). Requires srun --mpi=pmi2 with the full binary path. Use 6.4.2 instead unless you specifically need 6.2.1.
The old submission script looks like:
#!/bin/bash
#SBATCH -p sched_mit_rafagb
#SBATCH -C centos7
#SBATCH --nodes=1
#SBATCH -n 4
#SBATCH --cpus-per-task=1
#SBATCH --mem-per-cpu=4G
#SBATCH -t 0-06:00:00
#SBATCH -J vasp_job
#SBATCH --output=%j_%x.log
source ~/.bashrc
module purge
module use -a /nfs/rafagblab001/software/modulefiles
module load vasp/6.2.1
export OMP_NUM_THREADS=1
cd $SLURM_SUBMIT_DIR
srun --mpi=pmi2 /nfs/rafagblab001/software/vasp/vasp.6.2.1/bin/vasp_std
| Problem | Cause | Fix |
|---|---|---|
Illegal instruction | Binary compiled on different CPU arch | Recompile on the target node type (see Build section) |
cannot open shared object file: libscalapack.so | LD_LIBRARY_PATH not set | Use the module (module load vasp/6.4.2) which sets this |
Permission denied on POTCARs | Not in sched_mit_rafagb group | Ask admin to add you |
rsync hangs | Duo 2FA issue | Use scp instead |
| POTCAR mismatch | Species order wrong | Match POSCAR species order exactly |
MIT Engaging cluster reference. SLURM job scheduling, Coley group nodes, filesystem layout, Python environments, and common commands. Use when working with the Engaging HPC cluster.
Verifies that bibliography references exist and have accurate metadata. Use when the user asks to "check references", "verify a bibliography", "validate citations", "check for hallucinated citations", "check a .bib file", or mentions auditing a BibTeX file for accuracy. Looks up each entry against Crossref and arXiv, then spawns web-search subagents for non-indexed sources, flagging missing DOIs, wrong years, title mismatches, venue mismatches, and broken URLs.
Reference guide for the MolGPU shared GPU cluster. Machine specs, SSH setup, shared filesystem, conda/mamba, multi-GPU training tips, and common commands.
MassSpecGym benchmark reference — dataset schema, Python API, transforms, evaluation metrics, retrieval pipeline, and common patterns. Use when working with MassSpecGym data, models, or evaluation.
Machine learning workflows for molecular data. Use when building ML models with molecules, splitting datasets, selecting fingerprints for ML, avoiding data leakage, or doing scaffold-based train/test splits. Covers common pitfalls in molecular ML.
Python package and environment management using uv and mamba. Use when installing packages, creating virtual environments, setting up new projects, or managing dependencies. NOT for general Python coding questions.