spectral-norm benchmark N=5,500

Each chart bar shows how many times more Code, one ↓ spectral-norm program used, compared to the program that used least Code.

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
1.0Ruby JRuby 270.56263.82656,884292  34% 22% 18% 31%
1.0Ruby 6 min6 min8,972292  1% 100% 0% 0%
1.1Ruby #4 5 min5 min9,452326  1% 100% 0% 0%
1.1Ruby JRuby #4 251.58245.32660,988326  29% 35% 17% 24%
1.1Python 3 #6 12 min12 min6,464328  1% 0% 0% 100%
1.1Perl 200.90200.975,516333  0% 0% 0% 100%
1.2Perl #2 15 min15 min3,972343  85% 15% 0% 0%
1.3OCaml #2 15.6915.701,804377  42% 0% 0% 58%
1.3C gcc 20.6920.70468383  1% 0% 1% 100%
1.4PHP #2 7 min7 min7,244397  96% 0% 5% 0%
1.4Hack #2 68.9468.9756,400398  0% 100% 1% 1%
1.4Haskell GHC #2 42.1620.294,724403  100% 37% 37% 35%
1.4Scala 17.5917.5123,204404  1% 0% 1% 100%
1.4Go 15.7115.701,296411  1% 1% 0% 99%
1.4Pascal Free Pascal 15.7115.728423  1% 0% 100% 0%
1.5Python 3 #5 13 min209.7643,728437  98% 98% 98% 98%
1.5Racket 40.1640.1619,828446  33% 1% 67% 0%
1.5Python 3 #8 15 min15 min5,600449  0% 1% 100% 0%
1.5C++ g++ 20.7520.761,072452  1% 0% 100% 1%
1.6Dart 22.2021.9858,260457  1% 1% 1% 100%
1.6C# Mono 21.8721.8820,148459  1% 0% 100% 1%
1.6C gcc #3 15.753.95752463  100% 100% 100% 100%
1.7Erlang HiPE 56.7756.7817,920507  75% 25% 0% 0%
1.7Erlang 222.95223.0319,936507  90% 10% 0% 0%
1.8Fortran Intel #2 7.857.85788513  1% 0% 0% 100%
1.8Java  17.1217.0620,908514  99% 1% 1% 1%
1.8Racket #2 37.7937.7923,340532  0% 1% 100% 0%
1.8Go #3 15.703.961,816536  99% 99% 99% 99%
1.9Perl #4 183.4747.039,340551  98% 97% 98% 97%
1.9Fortran Intel 7.912.011,348568  99% 99% 99% 99%
1.9C gcc #5 7.891.98752569  99% 99% 100% 100%
2.1Lisp SBCL 21.2821.2912,752625  0% 1% 100% 0%
2.1Racket #3 20.665.4126,372627  95% 95% 98% 95%
2.2Fortran Intel #3 7.982.001,328638  100% 100% 100% 100%
2.3Go #2 15.774.111,296668  96% 96% 96% 96%
2.4Ada 2005 GNAT 21.7621.771,492710  1% 1% 100% 0%
2.5Scala #2 17.814.6325,420720  100% 95% 95% 95%
2.6Erlang HiPE #2 47.4812.0922,836747  99% 98% 98% 98%
2.6Erlang #2 145.6136.7816,772747  99% 99% 99% 99%
2.6Clojure #7 18.785.5860,924762  82% 81% 82% 93%
2.7Ruby JRuby #2 16 min250.73668,044776  98% 98% 98% 98%
2.8Clojure #6 18.765.5364,384808  83% 81% 81% 95%
2.8Ruby #3 6 min98.7042,696828  96% 96% 97% 97%
2.9Perl #3 15 min235.3610,404846  100% 100% 100% 100%
2.9F# Mono #2 22.347.3830,224852  73% 76% 82% 73%
3.0Lisp SBCL #3 15.914.1323,184883  97% 97% 96% 97%
3.1Lisp SBCL #2 15.914.1223,184906  96% 97% 97% 97%
3.1Clojure #8 18.305.2564,960918  84% 83% 98% 85%
3.2OCaml #3 16.824.783,512938  93% 91% 92% 90%
3.3Java  #2 17.804.5921,132950  97% 96% 99% 96%
3.4Haskell GHC #4 15.704.041,940984  97% 97% 97% 97%
3.6C++ g++ #5 10.092.544961044  100% 100% 100% 100%
3.6C++ g++ #6 10.092.544961044  99% 99% 100% 100%
3.6C# Mono #2 21.945.6422,8041063  97% 97% 98% 98%
3.9C gcc #4 10.652.684961139  100% 100% 100% 100%
4.1PHP #3 7 min113.8524,4081193  98% 97% 100% 98%
4.1Hack #3 73.7618.89188,2281195  98% 97% 98% 98%
4.4C++ g++ #8 10.682.684961278  100% 100% 100% 100%
5.8Ada 2005 GNAT #3 16.914.341,9241702  99% 99% 99% 94%
9.5Ada 2005 GNAT #4 7.882.011,9282762  99% 99% 99% 99%
Scala #4 Failed1006
Scala #3 Failed982
"wrong" (different) algorithm / less comparable programs
0.8Python 3 #2 11.0111.02489,356233
1.3Python 3 #3 96.4397.021,197,752379
2.3C gcc #2 2.332.34236,904669
4.4C++ g++ #7 8.492.144921283
4.6C++ g++ #2 7.921.994961330
missing benchmark programs
Rust No program

 spectral-norm benchmark : Eigenvalue using the power method

diff program output N = 100 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 calculate the spectral norm of an infinite matrix A, with entries a11=1, a12=1/2, a21=1/3, a13=1/4, a22=1/5, a31=1/6, etc

Each program must implement 4 separate functions / procedures / methods like the C# program.

For more information see challenge #3 in Eric W. Weisstein, "Hundred-Dollar, Hundred-Digit Challenge Problems" and "Spectral Norm".

From MathWorld--A Wolfram Web Resource.
http://mathworld.wolfram.com/Hundred-DollarHundred-DigitChallengeProblems.html
http://mathworld.wolfram.com/SpectralNorm.html

Thanks to Sebastien Loisel for this benchmark.

Revised BSD license

  Home   Conclusions   License   Play