Hi,
I am running a LAMMPS test on an Intel cluster.
For the same test, I can sucessfully finish runs uner 64 cores.
But when I run with 128/256 cores, I would got following error.
Could anyone give some help on this? Many thanks
PPPM initialization …
G vector (1/distance)= 0.248835
grid = 25 32 32
stencil order = 5
estimated absolute RMS force accuracy = 0.0355478
estimated relative force accuracy = 0.000107051
using double precision FFTs
3d grid and FFT values/proc = 972 112
Setting up run …
Fatal error in MPI_Allreduce: Message truncated, error stack:
MPI_Allreduce(1890)…: MPI_Allreduce(sbuf=0x7fff18285c60, rbuf=0x200dd30, count=1, MPI_INT, MPI_MAX, MPI_COMM_WORLD) failed
MPIR_Allreduce(294)…:
MPIR_Reduce(1039)…:
MPIR_Reduce_binomial(172)…:
MPIDI_CH3_PktHandler_EagerShortSend(356): Message from rank 1 and tag 11 truncated; 8 bytes received but buffer size is 4
Hi,
I am running a LAMMPS test on an Intel cluster.
For the same test, I can sucessfully finish runs uner 64 cores.
But when I run with 128/256 cores, I would got following error.
Could anyone give some help on this? Many thanks
no. not without additional information.
please let us know:
- which version of lammps are you using?
- which input are you running? is it a specific "test", or one of the
benchmarks/examples shipped with LAMMPS?
- what platform (OS, CPU, compiler, MPI library) exactly are you running on?
- can you reproduce the same error on a different machine under the
same conditions.
without this information you are completely on your own.
you may be anyway, since the error message you copy hints at a bug in
the MPI library, but nobody can say that for certain without external
verification.
I was using latest verson from website.
And the input was a specfic test provide by others, not shipped from LAMMPS.
OS is Redhat AS6 and CPU is Intel Xeon E5
Both compiler and MPI are from Intel.
I will try different MPI and different machines soon.