Win64 on Intel Processor
Absoft 15.0.2 |
Absoft(AP) 15.0.2 |
FTN95 7.20 |
Intel 16.0 |
Intel(AP) 16.0 |
Lahey-GNU Lassen-4.93 |
NAG 6.0 |
PGI 15.7 |
|
AC | 5.25 | 5.53 | 10.75 | 4.65 | 4.67 | 6.32 | 7.13 | 6.49 |
AERMOD | 12.94 | 13.86 | 24.75 | 12.19 | 12.53 | 18.97 | 18.67 | 12.17 |
AIR | 3.76 | 2.72 | 7.35 | 2.97 | 2.51 | 4.80 | 5.18 | 5.32 |
CAPACITA | 21.05 | 21.60 | 43.75 | 18.50 | 18.79 | 20.30 | 22.63 | 18.48 |
CHANNEL2 | 85.03 | 44.11 | 230.58 | 93.59 | 47.12 | 100.26 | 132.79 | 93.84 |
DODUC | 19.36 | 20.94 | 34.58 | 15.49 | 15.37 | 21.39 | 22.63 | 19.54 |
FATIGUE2 | 60.79 | 68.67 | 280.52 | 57.04 | 55.07 | 33.55 | 89.22 | 127.84 |
GAS_DYN2 | 66.90 | 35.26 | 478.45 | 69.07 | 39.73 | 195.67 | 178.35 | 124.36 |
INDUCT2 | 80.03 | 116.83 | 409.46 | 72.10 | 54.35 | 74.72 | 124.43 | 126.41 |
LINPK | 6.11 | 5.40 | 6.24 | 5.04 | 5.22 | 5.31 | 7.52 | 7.11 |
MDBX | 9.36 | 9.31 | 14.29 | 6.61 | 4.89 | 8.15 | 8.75 | 9.23 |
MP_PROP_DESIGN | 121.86 | 22.31 | 600.68 | 64.47 | 11.34 | 195.94 | 176.82 | 96.20 |
NF | 8.90 | 8.94 | 16.49 | 8.22 | 7.89 | 8.79 | 9.65 | 10.16 |
PROTEIN | 21.01 | 21.18 | 38.19 | 22.41 | 22.51 | 20.60 | 19.52 | 23.12 |
RNFLOW | 15.20 | 19.24 | 22.27 | 13.26 | 11.28 | 16.02 | 16.98 | 17.94 |
TEST_FPU2 | 63.77 | 65.27 | 132.94 | 46.26 | 41.76 | 69.23 | 82.09 | 51.71 |
TFFT2 | 48.80 | 49.85 | 61.13 | 48.13 | 49.34 | 44.84 | 53.60 | 51.98 |
Geometric Mean | 23.31 | 20.21 | 53.34 | 20.16 | 15.96 | 25.71 | 30.29 | 26.92 |
Compiler Switches | |
Absoft | af90 -m64 -O4 -speed_math=10 -fast_math -march=core -xINTEGER -stack:0x30000000 |
Absoft (autoparallel) | -m64 -mavx -O5 -speed_math=10 -fast_math -march=core -xINTEGER -stack:0×30000000 |
FTN95 | ftn95 /p6 /optimize (slink was used to increase the stack size) |
Intel | ifort /fast /link /stack:64000000 |
Intel (autoparallel) | ifort /fast /Qparallel /link /stack:64000000 |
Lahey-GNU | lgf -64 -ofast -unroll -wpo -stack 64000000 |
NAG | nagfor -abi=64 -O4 -s -v -V |
PGI | pgf90 -V -fastsse -Munroll=n:4 -Mipa=fast,inline |
Notes |
All figures are execution times in seconds – measured on a machine with 2 x Xeon E5-2643 3.30GHz quad core processors, running at stock speed with hyperthreading disabled, 64 GBytes memory, and running 64-bit Windows 7 64-bit. Each figure is the average over at least 10 runs (many more for some). Measurement error is typically <1%. Green cells highlight figures within 10% of the fastest. Red cells indicate figures which are more than 150% of the fastest. So far as possible, we have used the compiler switches which give the best overall results. We have not attempted to tune individual benchmarks, and, in particular cases, different switch settings may give better results. Where possible, compiler switches were set to generate 64 bit executables. The settings used for the Absoft and Intel compilers enable autoparallelization. Autoparallelization settings are not used on any other compilers because we found that they produced no significant performance benefits on this benchmark set. Thanks are due to Jos Bergervoet for permission to use his CAPACITA benchmark, to Quetzal Associates for permission to use their CHANNEL, FATIGUE, GAS_DYN, INDUCT, PROTEIN and RNFLOW benchmarks, to David Frank for his TEST_FPU benchmark, to Anthony Falzone for the use of MP_PROP_DESIGN, and to Ted Addison of McVehil-Monnett Associates for permission to use AERMOD, an air quality model used by the US Environmental Protection Agency. All the benchmarks have been modified slightly to fit into our benchmarking harness. |