开发者
资源
运行vasp出现openMPI报错:mca_coll_hcoll_comm_query failed
运行vasp出现openMPI报错:mca_coll_hcoll_comm_query failed
发表于2023/07/04
1260
背景:使用openmpi编译VASP,运行直接报错,编译命令如下:

./configure --prefix=/path/to/OPENMPI --enable-pretty-print-stacktrace --enable-orterun-prefix-by-default --with-knem=/opt/knem-1.1.3.90mlnx1/ --with-hcoll=/opt/mellanox/hcoll/ --with-cma --with-ucx --enable-mpi1-compatibility CC=gcc CXX=g++ FC=gfortran

36b9be0d44a249ec1c857961ce91942e_897x251.png%40900-0-90-f.png

根因分析:

hcoll是HPCX特有的模块,openmpi不支持该模块,需要屏蔽

70947d5467bebc5f717a95e1b344adb3_707x613.png%40900-0-90-f.png

解决方案:

a.重新编译openmpi,将hcoll编译选项去除,或者

b.运行命令默认不走hcoll协议:

mpirun --allow-run-as-root -np 96 -mca coll ^hcoll -x OMP_NUM_THREADS=1 vasp_std 2>&1 | tee -a vasp.out

收藏举报
Level 1
0
帖子
0
粉丝
0
获赞