Lennard Jones Benchmark - Fixed-size of 32K atoms
Lennard Jones Benchmark - Scaled-size of 32K atoms/processor
LAMMPS bio-membrane Benchmark - Fixed-size of 7K atoms
LAMMPS bio-membrane Benchmark - Scaled-size of 7K atoms/processor
This data is for running two different molecular dynamics (MD) benchmarks for both fixed-size and scaled-size problems on the Intel Tflops and CPlant siberia machines.
The 1st 2 graphs are for Lennard-Jones MD simulations. A 3-d box of atoms is simulated at liquid density using a standard 2.5 sigma cutoff. The simulation box is partitioned across processors using a spatial-decomposition algorithm -- look here for more info.
The fixed-size LJ problem results are for a 32,000 atom system. The scaled-size results are for system sizes of 32,000 atoms/processor -- i.e. on one processor a 32,000 atom simulation was run while on 256 processors, 8.2 million atoms were simulated. For both problems, one-processor timings on both machines are given in the lower left corner of the graphs.
The 2nd set of 2 graphs are for running a full-blown molecular dynamics simulation with the LAMMPS code of a lipid bilayer solvated by water. A 12 Angstrom cutoff was used for the van der Waals forces; long-range Coulombic forces were solved for using the particle-mesh Ewald method.
The 1st plot of fixed-size results is for a 7134-atom bilayer. The 2nd plot is scaled-size results for 7134 atoms/processor -- i.e. on one processor a 7134-atom simulation was run while on 1024 processors, 7.1 million atoms were simulated. Again, one processor timings for both machines (per MD timestep) are shown at the lower left of the plots. On the scaled-size problem, the CPlant machine does well until 256 processors when the parallel FFTs take a bit hit in parallel efficiency.
More info about these and other Cplant vs Tflop benchmarks can be found on this page. A paper written by Ron Brightwell and myself (Steve Plimpton) is also downloadable from there.