n-body benchmark hypotheses non fingo
↓ Double-precision N-body simulation, N=20,000,000.
Which programs used least Code? Which programs use highly optimised assembly code libraries? Which programs make use of all the processor cores?
| sort | sort | sort | sort | |||
|---|---|---|---|---|---|---|
| × | Program & Logs | CPU secs | Elapsed secs | Memory KB | Code B | ~ CPU Load |
| 1.0 | Oberon-2 OO2C | 14.60 | 1,696 | 1465 | ||
| 1.0 | C++ GNU g++ #2 | 14.62 | 932 | 1705 | ||
| 1.0 | Java 6 -server #2 | 14.75 | 11,524 | 1424 | ||
| 1.0 | Eiffel SmartEiffel | 14.81 | 520 | 1465 | ||
| 1.0 | C GNU gcc | 14.84 | 396 | 1173 | ||
| 1.0 | Java 6 -Xms64m #2 | 14.84 | 11,652 | 1424 | ||
| 1.1 | C++ GNU g++ #3 | 15.49 | 804 | 1344 | ||
| 1.1 | Java 1.4 -server | 15.75 | 10,296 | 1404 | ||
| 1.1 | C++ GNU g++ | 15.84 | 848 | 1349 | ||
| 1.1 | Java 6 -server | 15.93 | 12,012 | 1430 | ||
| 1.1 | Java 6 -Xms64m | 16.09 | 12,116 | 1430 | ||
| 1.1 | Java GNU gcj | 16.14 | 15,588 | 1404 | ||
| 1.1 | Fortran Intel #2 | 16.36 | 440 | 1244 | ||
| 1.1 | Scala | 16.43 | 20,796 | 1420 | ||
| 1.1 | C Intel | 16.60 | 444 | 1173 | ||
| 1.1 | Fortran Intel | 16.68 | 440 | 1389 | ||
| 1.2 | Pascal Free Pascal | 16.91 | 272 | 1308 | ||
| 1.2 | Nice #2 | 17.07 | 12,064 | 1313 | ||
| 1.2 | C++ Intel #3 | 17.16 | 236 | 1344 | ||
| 1.2 | Clean | 17.28 | 520 | 1587 | ||
| 1.2 | C GNU gcc #3 | 17.45 | 396 | 1315 | ||
| 1.2 | C Intel #3 | 17.74 | 440 | 1315 | ||
| 1.2 | Lisaac | 17.77 | 480 | 1615 | ||
| 1.3 | Java 6 -client #2 | 18.79 | 10,088 | 1424 | ||
| 1.4 | Java 6 -client | 20.39 | 10,280 | 1430 | ||
| 1.4 | Fortran G95 | 20.40 | 540 | 1389 | ||
| 1.4 | D Digital Mars | 20.67 | 688 | 1308 | ||
| 1.4 | CAL | 20.93 | 14,364 | 2356 | ||
| 1.5 | SML SML/NJ | 21.51 | 2,624 | 1314 | ||
| 1.5 | Lisp SBCL | 22.53 | 16,952 | 1363 | ||
| 1.6 | C# Mono #2 | 23.20 | 5,072 | 1410 | ||
| 1.6 | C++ Intel | 23.93 | 420 | 1349 | ||
| 1.8 | D Digital Mars #2 | 25.91 | 680 | 1193 | ||
| 1.8 | Ada 2005 GNAT | 26.38 | 480 | 1608 | ||
| 1.9 | SML MLton #2 | 27.27 | 740 | 1327 | ||
| 1.9 | Nice | 28.33 | 12,640 | 1304 | ||
| 2.0 | C# Mono | 29.39 | 5,060 | 1403 | ||
| 2.3 | BASIC FreeBASIC #4 | 33.84 | 772 | 1426 | ||
| 2.4 | Zonnon Mono | 34.61 | 5,168 | 1524 | ||
| 3.3 | Haskell GHC | 48.72 | 1,648 | 1687 | ||
| 3.5 | Forth bigForth | 51.00 | 720 | 1097 | ||
| 3.7 | OCaml | 54.00 | 644 | 1239 | ||
| 7.2 | F# Mono | 105.17 | 7,636 | 1364 | ||
| 8.7 | Lua LuaJIT #2 | 127.43 | 956 | 1265 | ||
| 9.6 | C++ Intel #2 | 140.49 | 444 | 1705 | ||
| 10 | Erlang HiPE #3 | 147.59 | 4,736 | 1399 | ||
| 11 | Fortran G95 #2 | 161.99 | 544 | 1244 | ||
| 15 | Scheme Ikarus | 225.32 | 28,016 | 1598 | ||
| 21 | Scheme Chicken | 5 min | 1,988 | 1394 | ||
| 23 | Scheme Chicken #3 | 5 min | 2,300 | 1337 | ||
| 27 | Mercury | 6 min | 5,028 | 1781 | ||
| 27 | Forth GNU GForth | 6 min | 888 | 1304 | ||
| 27 | Java 6 -Xint #2 | 6 min | 9,556 | 1424 | ||
| 33 | Scheme PLT #2 | 7 min | 8,036 | 1423 | ||
| 33 | Smalltalk VisualWorks | 8 min | 12,720 | 1652 | ||
| 33 | Java 6 -Xint | 8 min | 9,692 | 1430 | ||
| 34 | Python Psyco #4 | 8 min | 3,644 | 1185 | ||
| 36 | Scheme PLT | 8 min | 7,096 | 1544 | ||
| 40 | Lua #2 | 9 min | 872 | 1265 | ||
| 44 | Python Psyco #2 | 10 min | 3,256 | 1044 | ||
| 86 | Mozart/Oz | 20 min | 4,344 | 1444 | ||
| 96 | Pike | 23 min | 3,416 | 1260 | ||
| 99 | Prolog YAP #2 | 24 min | 22,800 | 1455 | ||
| 105 | Groovy | 25 min | 55,020 | 1214 | ||
| 116 | Perl | 28 min | 1,936 | 1140 | ||
| 121 | Ruby JRuby #2 | 29 min | 41,964 | 1137 | ||
| 135 | JavaScript Rhino | 32 min | 13,636 | 1287 | ||
| 138 | Python CPython #2 | 33 min | 2,440 | 1027 | ||
| 140 | Icon | 34 min | 1,228 | 1287 | ||
| 145 | Python IronPython | 35 min | 16,704 | 1072 | ||
| 148 | Python CPython | 36 min | 2,440 | 1072 | ||
| 163 | PHP | 39 min | 5,636 | 1289 | ||
| 166 | Smalltalk GNU | 40 min | 4,288 | 1657 | ||
| 169 | Python IronPython #2 | 41 min | 16,792 | 1027 | ||
| 186 | Ruby 1.9 #2 | 45 min | 2,468 | 1137 | ||
| 236 | JavaScript SpiderMonkey | 57 min | 66,452 | 1287 | ||
| 319 | Prolog SWI #2 | 1h 17 min | 2,488 | 1573 | ||
| 320 | Tcl #2 | 1h 17 min | 1,960 | 1454 | ||
| 332 | Ruby MRI #2 | 1h 20 min | 1,712 | 1137 | ||
| 595 | Prolog SWI | 2h 24 min | 2,524 | 1586 | ||
| 916 | Rebol | 3h 42 min | 2,780 | 1164 | ||
| C CINT | Timed Out | 1173 | ||||
| Io | Failed | 1106 | ||||
| Scala #2 | Failed | 1430 | ||||
| interesting alternative programs | ||||||
| 0.9 | C GNU gcc #2 | 13.86 | 0.00 | 396 | 1186 | |
| 1.1 | C Intel #2 | 15.90 | 0.00 | 444 | 1186 | |
| 13 | Smalltalk Squeak | 193.64 | 0.00 | 20,084 | 1652 | |
| missing programs | ||||||
| Smalltalk Squeak | No program | |||||
n-body benchmark : Double-precision N-body simulation
ndiff program output N = 1000 with this output file to check your program is correct before contributing.
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.
