n-body benchmark N=50,000,000

Each chart bar shows how many times more Memory, one ↓ n-body program used, compared to the program that used least Memory.

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.

    sortsort sort
  ×   Program Source Code CPU secs Elapsed secs Memory KB Code B ≈ CPU Load
1.0Pascal Free Pascal 34.6934.7081308  1% 0% 100% 0%
1.0Pascal Free Pascal #2 37.2037.2281322  0% 1% 1% 100%
1.0Pascal Free Pascal #3 47.1847.1981418  0% 1% 1% 100%
65Fortran Intel #3 25.2225.235161299  0% 1% 100% 0%
65Fortran Intel #5 10.7510.765201659  1% 0% 100% 1%
65Fortran Intel #4 22.6022.615201172  1% 0% 100% 1%
65Fortran Intel 20.9020.915201389  1% 0% 0% 100%
66Fortran Intel #2 20.5320.545241496  0% 1% 1% 100%
73C gcc #5 21.0721.085801429  1% 0% 0% 100%
75OCaml 75.2375.256001239  1% 0% 0% 100%
79C++ g++ #4 20.3020.316281428  1% 0% 0% 100%
90C gcc 20.1920.197161173  0% 0% 100% 0%
91C gcc #2 25.2725.287281263  100% 0% 1% 0%
92C gcc #4 10.3810.397361490  0% 100% 1% 1%
95Go 23.1423.137601310  0% 1% 100% 1%
98C gcc #3 21.3021.317841208  1% 100% 0% 0%
99C gcc #6 20.5920.607881180  1% 0% 100% 0%
100C++ g++ #8 9.139.147961544  1% 100% 0% 0%
101C++ g++ 20.5720.588041659  0% 100% 1% 0%
101C++ g++ #6 20.1120.128081668  1% 2% 1% 100%
110C++ g++ #7 9.129.138761545  0% 0% 100% 1%
151C++ g++ #5 12.4812.491,2081749  100% 0% 0% 1%
154C++ g++ #3 19.8719.881,2281763  1% 0% 0% 100%
204Ada 2005 GNAT #5 18.0418.051,6322436  1% 0% 0% 100%
214Ada 2005 GNAT #3 25.5925.601,7121740  0% 0% 100% 0%
216Ada 2005 GNAT 24.8224.831,7241608  1% 0% 0% 100%
319Perl #2 20 min20 min2,5521401  1% 0% 0% 100%
324PHP #3 11 min11 min2,5921082  0% 1% 1% 100%
456Haskell GHC #2 27.1026.373,6441874  100% 2% 1% 2%
530Python 3 16 min16 min4,2401181  0% 0% 100% 1%
636Ruby #2 24 min24 min5,0841137  22% 0% 79% 1%
753Rust 24.5724.586,0241474  100% 0% 0% 1%
760Python 3 #2 16 min16 min6,0801228  6% 100% 2% 7%
982Lisp SBCL 46.0246.037,8521363  0% 0% 100% 0%
982Lisp SBCL #2 38.4138.437,8521398  0% 1% 100% 0%
1,550Haskell GHC 30.9329.1712,4001706  2% 100% 2% 2%
2,616Racket #2 55.3255.3320,9281689  1% 43% 58% 0%
2,684Dart #2 34.0233.7421,4721274  1% 55% 45% 1%
2,694Dart #3 38.6738.3321,5521420  1% 1% 1% 100%
3,020Java  #2 24.3424.3124,1561424  1% 1% 1% 100%
3,472Erlang HiPE #3 269.21269.3027,7761399  53% 1% 1% 47%
3,604Scala 23.4523.4028,8321358  1% 1% 100% 1%
4,567Racket 116.11116.1036,5361496  0% 100% 1% 1%
4,844C# Mono #2 23.1623.1738,7521410  1% 0% 0% 100%
4,928C# Mono 38.5838.5939,4241403  100% 0% 0% 1%
5,858F# Mono 25.9926.0046,8601259  1% 0% 100% 1%
6,785Clojure 34.6033.2654,2762324  98% 2% 2% 3%
6,835Clojure #2 38.1036.8354,6801440  98% 2% 3% 1%
81,909Ruby JRuby #2 7 min7 min655,2681137  36% 21% 24% 23%
C# Mono #8 Failed1343
C# Mono #5 Failed1485
C# Mono #3 Failed1305
C# Mono #7 Failed1539
C# Mono #6 Failed1289
C# Mono #4 Failed1461
Java  Failed1424
"wrong" (different) algorithm / less comparable programs
151C++ g++ #2 12.2912.301,2082288

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

You can write your own program for this task and contribute to the benchmarks game by following these general instructions.

More specifically:

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.

Each program should model the orbits of Jovian planets, using the same simple symplectic-integrator - see the Java program.

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