Chapel versus Intel Fortran fastest performance
How the programs are written matters! Always look at the source code.
If the fastest programs are * hand-written vector instructions, does the host language matter? You might be more interested in the less optimised programs — more seconds, less gz source code.
fannkuch-redux | ||||
---|---|---|---|---|
source | secs | mem | gz | cpu secs |
Chapel #2 | 8.42 | 19,620 | 737 | 33.54 |
Intel Fortran #3 | 9.78 | 19,648 | 1164 | 39.00 |
Intel Fortran | 40.33 | 19,648 | 608 | 40.33 |
n-body | ||||
source | secs | mem | gz | cpu secs |
Intel Fortran #5 | 3.39 | 19,648 | 1669 | 3.39 |
Chapel #3 | 3.56 | 19,620 | 983 | 3.57 |
Intel Fortran #2 | 4.46 | 19,776 | 1507 | 4.46 |
Intel Fortran | 4.55 | 19,648 | 1400 | 4.55 |
Chapel #4 | 4.64 | 19,620 | 1027 | 4.64 |
Intel Fortran #3 | 4.71 | 19,648 | 1311 | 4.71 |
Intel Fortran #6 | 5.01 | 19,648 | 1530 | 5.01 |
Chapel #2 | 5.69 | 19,620 | 977 | 5.70 |
Intel Fortran #4 | ||||
spectral-norm | ||||
source | secs | mem | gz | cpu secs |
Intel Fortran #2 | 2.67 | 19,648 | 520 | 2.66 |
Chapel #2 | 0.73 | 19,620 | 348 | 2.88 |
Intel Fortran #3 | 0.73 | 19,648 | 644 | 2.88 |
Chapel | 1.46 | 19,620 | 322 | 5.77 |
mandelbrot | ||||
source | secs | mem | gz | cpu secs |
Chapel #3 | 1.14 | 37,780 | 582 | 4.51 |
Intel Fortran #6 | 2.71 | 79,488 | 669 | 8.55 |
Intel Fortran #7 | 4.08 | 79,508 | 965 | 16.18 |
Intel Fortran #4 | 9.85 | 79,744 | 626 | 25.23 |
Chapel | 6.64 | 37,820 | 447 | 26.47 |
Intel Fortran #5 | 32.58 | 43,648 | 528 | 32.58 |
fasta | ||||
source | secs | mem | gz | cpu secs |
Chapel #6 | 0.79 | 19,620 | 1108 | 0.80 |
Intel Fortran #4 | 1.34 | 19,648 | 1355 | 1.34 |
Chapel #5 | 0.97 | 19,620 | 1391 | 3.53 |
Chapel #3 | 4.03 | 19,620 | 1219 | 4.04 |
Intel Fortran | 5.83 | 19,648 | 1193 | 5.83 |
Intel Fortran #3 | 6.19 | 19,648 | 1217 | 6.19 |
k-nucleotide | ||||
source | secs | mem | gz | cpu secs |
Chapel #4 | 5.15 | 202,548 | 1074 | 17.53 |
Chapel #3 | 6.14 | 203,208 | 1044 | 21.26 |
Intel Fortran | ||||
reverse-complement | ||||
source | secs | mem | gz | cpu secs |
Chapel #9 | 0.98 | 1,057,244 | 1260 | 2.81 |
Chapel #3 | 1.69 | 2,133,608 | 709 | 3.07 |
Chapel #5 | 1.93 | 1,905,304 | 554 | 3.28 |
Intel Fortran | 3.33 | 986,304 | 1026 | 3.33 |
Intel Fortran #2 | 42.90 | 769,748 | 778 | 42.89 |
binary-trees | ||||
source | secs | mem | gz | cpu secs |
Intel Fortran #2 | 2.22 | 140,544 | 1155 | 6.86 |
Chapel #5 | 2.04 | 1,775,900 | 646 | 7.26 |
Chapel #3 | 8.57 | 303,104 | 494 | 25.97 |
Chapel #4 | 7.22 | 480,044 | 489 | 26.54 |
Intel Fortran | 32.59 | 525,824 | 780 | 32.59 |
pidigits | ||||
source | secs | mem | gz | cpu secs |
Chapel #2 | 0.75 | 19,620 | 423 | 0.76 |
* Chapel #4 | 0.76 | 19,620 | 515 | 0.76 |
Intel Fortran #4 | 0.88 | 19,648 | 987 | 0.88 |
regex-redux | ||||
source | secs | mem | gz | cpu secs |
Intel Fortran | 0.96 | 179,724 | 2229 | 3.09 |
Chapel #3 | 3.11 | 404,924 | 503 | 4.05 |
Chapel #2 | 3.54 | 405,032 | 498 | 4.48 |
Chapel | ||||
Intel Fortran |