fannkuch-redux benchmark N=12

Each chart bar shows how many times slower, one ↓ fannkuch-redux 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
1.0OCaml #4 0.0033.579,2681004  100% 100% 100% 100%
1.0OCaml #3 0.0026.4218,6961017  100% 100% 100% 100%
6,816C gcc #4 27.2627.273161183  0% 0% 1% 100%
6,818C++ g++ #7 27.2727.282841150  0% 100% 0% 1%
9,823Ada 2005 GNAT #3 39.299.871,6562100  100% 100% 100% 100%
11,994C gcc #3 47.9847.99316567  0% 100% 1% 1%
12,508C++ g++ #5 50.0312.881,0361440  99% 99% 100% 99%
13,119C++ g++ #4 52.4813.371,0361439  100% 100% 99% 95%
13,693C gcc #2 54.7714.243441557  91% 98% 100% 96%
14,470C gcc 57.8857.90316508  0% 100% 1% 1%
14,478C++ g++ #3 57.9157.93284593  1% 0% 0% 100%
15,311Lisp SBCL #4 61.2415.7926,2961518  92% 99% 100% 98%
15,425Lisp SBCL #5 61.7061.7222,940674  0% 0% 100% 1%
16,104C# Mono #2 64.4264.4318,580564  1% 0% 0% 100%
16,113Haskell GHC #3 64.4516.433,4521153  100% 100% 99% 94%
16,722Fortran Intel #3 66.8916.858,5081148  100% 100% 100% 100%
16,884Pascal Free Pascal 67.5416.936961018  100% 100% 100% 100%
17,011Java  68.0417.2515,6961282  99% 99% 97% 99%
17,102Fortran Intel 68.4168.43520590  100% 0% 0% 0%
17,350Lisp SBCL #3 69.4069.4216,068821  1% 0% 0% 100%
17,655Rust 70.6217.7137,720644  100% 100% 100% 99%
17,694F# Mono #4 70.7870.8019,104612  1% 100% 0% 0%
17,775OCaml #2 71.1071.12596473  0% 0% 0% 100%
18,472Scala #2 73.8918.7519,8201017  99% 99% 100% 97%
19,124Java  #2 76.5076.4616,164514  0% 99% 1% 1%
19,142Haskell GHC #5 76.5719.383,720834  100% 96% 100% 99%
20,496Dart #2 81.9881.4089,452495  1% 1% 1% 100%
21,864Haskell GHC #4 87.4684.031,932658  100% 2% 2% 2%
22,242C# Mono 88.9788.9918,576520  1% 1% 100% 1%
25,152Scala 100.61100.5518,988459  0% 1% 100% 0%
26,109Go 104.4426.19772900  100% 100% 100% 100%
27,018F# Mono #2 108.07108.1020,612548  1% 0% 100% 1%
28,188C# Mono #3 112.7528.6919,5321096  99% 98% 99% 98%
31,360F# Mono #3 125.4432.8623,416945  93% 97% 95% 97%
35,162OCaml 140.65140.69592524  0% 0% 0% 100%
37,224Clojure #3 148.9038.1048,5281491  97% 98% 98% 98%
42,425F# Mono 169.70169.6723,196551  29% 3% 1% 69%
46,387Lisp SBCL #2 185.55185.6835,636513  13% 2% 100% 25%
63,545Clojure #2 254.1874.0553,4441088  85% 86% 86% 84%
76,043Racket #2 5 min5 min16,876903  0% 0% 30% 70%
78,446Racket 5 min5 min15,364649  0% 1% 41% 60%
83,164Racket #3 5 min83.7717,1601096  99% 100% 99% 100%
101,163Erlang HiPE 6 min104.748,1801038  95% 99% 99% 94%
112,578Haskell GHC #2 7 min144.594,996808  79% 78% 78% 78%
144,650Haskell GHC 9 min7 min3,768553  71% 7% 7% 38%
222,727Ruby JRuby 14 min14 min620,384384  5% 41% 42% 22%
622,251Perl #2 41 min10 min6,292565  100% 100% 99% 100%
655,993Python 3 #6 43 min43 min4,200385  0% 0% 1% 100%
678,895PHP #2 45 min45 min2,544441  0% 0% 0% 100%
705,079Perl 47 min47 min1,752457  100% 0% 0% 0%
711,794PHP #3 47 min12 min10,5641150  91% 93% 99% 100%
769,754Ruby 51 min51 min5,288384  0% 0% 0% 100%
865,861PHP 57 min57 min2,576482  0% 98% 3% 0%
1,000,014Python 3 #2 1h 06 min16 min27,840797  99% 99% 98% 100%
1,097,826Python 3 1h 13 min18 min27,8881108  100% 99% 99% 98%
C++ g++ Failed1059
"wrong" (different) algorithm / less comparable programs
9,126C++ g++ #6 36.5036.52284894
10,227Java  #3 40.9110.4016,9921633
13,342Lisp SBCL 53.3713.5033,6401607
21,582C# Mono #4 86.3386.3618,844710
28,422C# Mono #5 113.6928.8521,2161400

 fannkuch-redux benchmark : Indexed-access to tiny integer-sequence

diff program output N = 7 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 N = 7 programs should generate these permutations (40KB) - which, incidentally, seem to be in the same order as permutations generated by the Tompkins-Paige algorithm, see pages 150-151 Permutation Generation Methods Robert Sedgewick.

The fannkuch benchmark is defined by programs in Performing Lisp Analysis of the FANNKUCH Benchmark, Kenneth R. Anderson and Duane Rettig.

Each program should

The conjecture is that this maximum count is approximated by n*log(n) when n goes to infinity.

FANNKUCH is an abbreviation for the German word Pfannkuchen, or pancakes, in analogy to flipping pancakes.


Thanks to Oleg Mazurov for insisting on a checksum and providing this helpful description of the approach he took -

Revised BSD license

  Home   Conclusions   License   Play