Careers | Phone Book | A - Z Index

November 2014 Results (2nd Order)

Here are the older, second order (operators.7pt.c) HPGMG-FV results as presented at the Supercomputing 2014 BoF. Each machine was allowed to use any amount of memory per node. Unfortuantely, due to scheduling and allocation limitations, some machines only evaluated a single per-node problem size. Similarly, 'Fraction of System' represents what fraction of the system was available for benchmarking.  In order to reproduce these results, users should use the Chebyshev smoother (-DUSE_CHEBY or --fv-smoother=cheby) and modify finite-volume/source/local.mk to use operators.7pt.c.  However, submission for new ranking must use the approach described for the fourth order.

HPGMG     HPGMG Fraction of Parallelization  DOF per Top500
Rank System Site DOF/s System MPI OMP Process Rank
1 Mira Argonne 7.21E+11 100% 49152 64 16M 5
2 K RIKEN 7.12E+11 73% 64000 8 2M 4
3 Edison NERSC 3.85E+11 100% 131072 1 4M 24
4 Titan (CPU-only) Oak Ridge 2.53E+11 88% 32768 8 16M 2
5 Stampede (CPU-only) TACC 1.49E+11 64% 8192 8 2M 7
6 Hopper NERSC 1.21E+11 86% 21952 6 2M 44
7 Piz Daint (CPU-only) CSCS 1.02E+11 78% 4096 8 18M 6
8 SuperMUC LRZ 7.13E+10 15% 2744 8 16M 14
9

Stampede (MIC-only)

TACC 2.16E+10 8% 512 180 16M 7
10 Peregrine (IVB-only) NREL 1.08E+10 18% 512 12 2M -
11 Carver NERSC 1.35E+09 5% 125 4 2M -
12 Babbage (MIC-only) NERSC 8.24E+08 30% 27 180 16M -