
* Cray Titan with the "SC16" benchmark

The 'host' version here is 2 MPI tasks x 8 threads 
cf Gray and Stratford 2016.

This version is about the same as Alan's trunk version appearing
in the paper, although exact distribution of time is not the same.


Currently Loaded Modulefiles:
  1) eswrap/1.3.3-1.020200.1278.0
  2) craype-network-gemini
  3) cray-mpich/7.4.0
  4) craype-interlagos
  5) lustredu/1.4
  6) xalt/0.5.3
  7) module_msg/0.1
  8) modulator/1.2.0
  9) hsi/5.0.2.p1
 10) DefApps
 11) cudatoolkit/7.5.18-1.0502.10743.2.1
 12) craype-accel-nvidia35
 13) intel/16.0.3.210
 14) craype/2.5.5
 15) cray-libsci/16.06.1
 16) udreg/2.3.2-1.0502.10518.2.17.gem
 17) ugni/6.0-1.0502.10863.8.28.gem
 18) pmi/5.0.9-1.0000.10911.175.4.gem
 19) dmapp/7.0.1-1.0502.11080.8.74.gem
 20) gni-headers/4.0-1.0502.10859.7.8.gem
 21) xpmem/0.1-2.0502.64982.5.3.gem
 22) dvs/2.5_0.9.0-1.0502.2188.1.113.gem
 23) alps/5.2.4-2.0502.9774.31.12.gem
 24) rca/1.0.0-2.0502.60530.1.63.gem
 25) atp/2.0.2
 26) PrgEnv-intel/5.2.82
