Hi all,
I am new to lammps. I want to run lammps for benchmarking a large infiniband based hpc cluster ( > 50 TF) installed at CRL, Pune, India. I have installed lammps and ran some of the benchmark problems given in the "bench" folder. But those problems run very fast. I want some really large problems of practical significance to run on this cluster. Can anyone provide me input file(s) for very large problem(s). The problem can be form any field. I will run the problem on our system and I will be happy to share the results.
Regards,
Chandan
The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments contained in it.
Contact your Administrator for further information.
I am new to lammps. I want to run lammps for benchmarking a large
infiniband based hpc cluster ( > 50 TF) installed at CRL, Pune, India. I
have installed lammps and ran some of the benchmark problems given in
the "bench" folder. But those problems run very fast. I want some
you can easily blow up many of the given inputs by using
the replicate command (see the lammps docs for more details).
really large problems of practical significance to run on this cluster.
Can anyone provide me input file(s) for very large problem(s). The
yes. i have a large number of benchmarks on coarse grained MD
on a large variety of capable machines and with up to over a
million coarse grain beads. unfortunately, those pair interaction
classes are still work in progress, so you'd have to use a hacked
version of LAMMPS. i also have some inputs on a CHARMM forcefield
simulation (with comparison to NAMD 2.6).
problem can be form any field. I will run the problem on our system
and I will be happy to share the results.
how fast do you need the inputs (and the code)? i'm currently
pretty busy with a school on biosimulations in bangalore and
cannot promise a quick delivery or much help until i'm back
in philly.
The benchmarks that use create_atom, can have the lattice
size modified to run as big a problem as you want. The ones
that read from a data file, can use the replicate command to
make the system as big as you want, like the in.*.scaled files
already do.
Whether any of these problems have "practical significance"
is another question ...