我正在尝试使用pyOM2的fortran-mpi组件进行海洋模拟。因此,我已经使用Python 3和f2py
编译了pyOM。这都是Veros项目的一部分。
当我使用fortran
在此处运行基准测试时,一切正常。当我在fortran-mpi
的{{1}}机器上使用AMD A10-7850K APU with Radeon(TM) R7 Graphics
运行它时,它就可以正常工作。
现在,我尝试在/proc/cpuinfo
机器上运行相同的文件和配置。在这里,我收到以下MPI错误:
Intel(R) Xeon(R) CPU E5-2698 v4 @ 2.20GHz
我已经尝试了running benchmark isoneutral_benchmark.py
current size: 980
fortran-mpi ... failed
--------------------------------------------------------------------------
No OpenFabrics connection schemes reported that they were able to be used on a specific port. As such, the openib BTL (OpenFabrics support) will be disabled for this port.
Local host: scigpu01
Local device: mlx5_3
Local port: 1
CPCs attempted: udcm
--------------------------------------------------------------------------
testing mpi routines
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 73 in communicator MPI_COMM_WORLD with errorcode 99.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
ERROR:
domain decompositon impossible in j-direction
choose other number of PEs in j-direction
global pe # 73 : in pe_decomposition
global pe # 73 aborting
... # From ERROR down it repeats 10-12 times
[scigpu01:58598] 79 more processes have sent help message help-mpi-btl-openib-cpc-base.txt / no cpcs for port
[scigpu01:58598] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
[scigpu01:58598] 16 more processes have sent help message help-mpi-api.txt / mpi-abort
环境变量的各种设置,例如OMPI_MCA_btl
或^openib,sm,self
,但是却无所适从。
我可以运行任何魔术命令以使其在我的Intel机器上正常工作吗?
谢谢