That is great news. I will create and mount a new partition with the
settings suggested in the article and try again.
Cheers
On Sat, Sep 4, 2021, 12:55 Wolfgang Bangerth wrote:
> On 9/3/21 7:49 PM, Nicholas Yue wrote:
> > The file system I am mounting via NFS is an ordinary Linux file system,
@Matthias yes, it wrote out all 200 *.pvtu and *.vtu file pair and they are
non-empty.
Seems it is only the check pointing that is having problem via MPI-IO
FYI, I have now also created and mounted a different storage and followed
the `noac` NFS setting and have been able to run the full simulati
> I tried commenting out the call to check-pointing as you suggested and was
> able to run the code on 4 node (each with 4 cores) and it finished very
> quickly.
Did it write out vtu files during this run? (They are written using MPI-IO)
Best,
Matthias
--
The deal.II project is located at h
On 9/3/21 7:49 PM, Nicholas Yue wrote:
The file system I am mounting via NFS is an ordinary Linux file system, it is
not a HPC parallel filesystems like Lustre or anything like them.
I tried commenting out the call to check-pointing as you suggested and was
able to run the code on 4 node (each
The file system I am mounting via NFS is an ordinary Linux file system, it
is not a HPC parallel filesystems like Lustre or anything like them.
I tried commenting out the call to check-pointing as you suggested and was
able to run the code on 4 node (each with 4 cores) and it finished very
quic
Hi Nicholas,
On Fri, Sep 3, 2021, at 12:49 CDT, Nicholas Yue wrote:
> Hi
>
> It seems to be consistently failing when writing the checkpoint file(s)
>
> Are there special flags I need to setup up for some form of parallel IO
> that may be happening ?
> [...]
> Additional information:
>
On 9/3/21 11:49 AM, Nicholas Yue wrote:
An error occurred in line <1412> of file
in function
void dealii::parallel::DistributedTriangulationBasespacedim>::DataTransfer::save(unsigned int, unsigned int, const string&) const
[with int dim = 2; int spacedim = 2; std::__cxx11::string =
std::
Hi
I have build deal.ii on the Nvidia Jetson Nano cluster from Picocluster
It has 5 nodes (pc[0-4], I am using pc0 as the head/login node to launch
the application
The executable is in an NFS filesystem to share between all the nodes
The following works
mpirun --host pc1 --mca btl_tcp_if_incl
