n-body benchmark N=5,000,000

Each chart bar shows how many times slower, one ↓ n-body program was, compared to the fastest program.

These are not the only programs that could be written. These are not the only compilers and interpreters. These are not the only programming languages.

Column × shows how many times more each program used compared to the benchmark program that used least.

     sortsortsort
  ×   Program Source Code CPU secs Elapsed secs Memory KB Code B ≈ CPU Load
1.0PyPy 3 4.604.8268,9401315  3% 99% 5% 3%
1.1PyPy 3 4.854.8868,8401315  100% 16% 13% 10%
1.1PyPy 3 4.864.8968,8361315  15% 14% 95% 9%
1.1PyPy 2 4.944.9775,2361337  12% 16% 17% 100%
1.1PyPy 2 5.105.2175,0401337  20% 55% 21% 68%
1.1PyPy 2 5.185.3875,2441337  31% 77% 17% 38%
1.5Graal 6.756.60406,5601315  13% 80% 11% 7%
1.5Graal 6.806.69406,5481315  22% 1% 83% 6%
1.5Graal 6.876.81406,5641315  46% 4% 58% 4%
2.2Cython 10.0110.028,4201392  2% 100% 1% 2%
2.2Cython 10.0210.038,3441392  3% 1% 100% 2%
2.2Cython 10.0910.108,6241392  1% 1% 2% 100%
6.4Pyston 29.4429.4526,5961337  0% 53% 0% 47%
6.4Pyston 29.4929.4926,6481337  0% 0% 100% 0%
6.8Pyston 31.0531.0626,7241337  2% 0% 98% 0%
8.5Nuitka 38.9739.019,7641315  62% 1% 39% 1%
8.5Nuitka 39.0239.059,6721315  1% 39% 1% 62%
8.6Nuitka 39.4539.489,8321315  1% 1% 1% 100%
9.7Python 3 44.7145.218,8281315  37% 30% 44% 11%
10Python development version 45.6345.668,0001315  7% 9% 2% 93%
10Python development version 45.7345.768,1161315  7% 8% 2% 94%
10Python 3 45.7445.888,9841315  48% 7% 55% 2%
10Python development version 46.4546.487,9561315  6% 20% 1% 81%
10Python 3 47.6449.448,9321315  36% 79% 17% 17%
11Python 2 48.9348.956,7401337  1% 0% 0% 100%
11Python 2 50.9450.996,6161337  2% 17% 80% 6%
12Python 2 53.1753.336,6921337  30% 28% 64% 20%
12IronPython 55.0055.1965,2081337  69% 7% 11% 12%
12IronPython 55.2154.6763,0401337  0% 66% 0% 34%
12IronPython 56.3756.6963,3241337  30% 0% 68% 0%
13Numba 58.4358.4497,4921333  0% 100% 0% 0%
13Numba 59.6259.6297,3761333  0% 26% 0% 74%
13Numba 59.7460.0196,7361333  0% 0% 100% 0%
16Jython 71.2367.65297,4361337  28% 28% 24% 26%
16Jython 72.5868.57297,2521337  34% 21% 30% 21%
16Jython 74.2070.26299,0881337  20% 25% 33% 28%
23MicroPython 104.84109.374,4721315  1% 97% 2% 4%
23MicroPython 104.92107.864,4841315  32% 65% 6% 2%
23MicroPython 105.20109.114,4001315  7% 84% 14% 4%
missing benchmark programs
Shedskin No program
Grumpy No program
RustPython No program

 n-body benchmark : Double-precision N-body simulation

ndiff -abserr 1.0e-8 program output N = 1000 with this output file to check your program is correct before contributing.

We are trying to show the performance of various programming language implementations - so we ask that contributed programs not only give the correct result, but also use the same algorithm to calculate that result.

For background information see N-body problem. Useful symplectic integrators are freely available, for example the HNBody Symplectic Integration Package.

Thanks to Mark C. Lewis for this benchmark.

Revised BSD license

  Home   Conclusions   License   Play