HPC多节点并行计算错误
在安装并行计算是,针对MPICH3.3.2 的测试例子,用mpirun多节点并行机计算测试成功,但是并行计算wrf.exe时始终跑一会就中断了,报一以下的错误:
d01 2020-08-14_00:00:00 grid spacing, dt, time_step_sound= 9000.000 27.00000 4
d01 2020-08-14_00:00:00 call rk_step_prep
d01 2020-08-14_00:00:00 calling inc/HALO_EM_A_inline.inc
Fatal error in PMPI_Wait: Unknown error class, error stack:
PMPI_Wait(203)........................: MPI_Wait(request=0x4de87f4, status=0x7fff5ad13c20) failed
MPIR_Wait_impl(100)...................:
MPIDU_Complete_posted_with_error(1137): Process failed
有没有哪位大神可以看看如何解决吗?试了各种方法,换版本,修改ulmit参数配置等就是不行