spectral-norm benchmark N=5,500
Each chart bar shows how many times slower, one ↓ spectral-norm 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.
| sort | sort | sort | ||||
| × | Program Source Code | CPU secs | Elapsed secs | Memory KB | Code B | ≈ CPU Load |
|---|---|---|---|---|---|---|
| 1.0 | Fortran Intel #2 | 7.85 | 7.85 | 780 | 513 | 0% 1% 1% 100% |
| 1.0 | Fortran Intel #3 | 7.85 | 7.86 | 1,276 | 638 | 0% 0% 0% 100% |
| 1.0 | C gcc #5 | 7.86 | 7.86 | 644 | 569 | 0% 0% 0% 100% |
| 1.0 | Fortran Intel | 7.86 | 7.87 | 1,300 | 568 | 0% 1% 0% 100% |
| 1.0 | C++ g++ #5 | 7.92 | 7.93 | 924 | 1044 | 0% 0% 0% 100% |
| 1.0 | C++ g++ #6 | 7.92 | 7.93 | 924 | 1044 | 1% 0% 1% 100% |
| 1.2 | Ada 2005 GNAT #4 | 9.25 | 9.27 | 1,900 | 2762 | 0% 1% 0% 100% |
| 1.3 | C gcc #4 | 9.87 | 9.88 | 644 | 1139 | 0% 0% 0% 100% |
| 1.4 | ATS #4 | 10.60 | 10.60 | 608 | 2507 | 1% 0% 0% 100% |
| 1.7 | C++ g++ #8 | 13.54 | 13.55 | 644 | 1278 | 0% 0% 0% 100% |
| 2.0 | JavaScript V8 #2 | 15.70 | 15.71 | 7,220 | 328 | 0% 1% 1% 100% |
| 2.0 | JavaScript V8 #3 | 15.70 | 15.71 | 7,412 | 373 | 0% 0% 0% 100% |
| 2.0 | JavaScript V8 | 15.70 | 15.71 | 7,504 | 311 | 0% 0% 1% 100% |
| 2.0 | Ada 2005 GNAT #3 | 15.70 | 15.72 | 1,636 | 1702 | 0% 0% 0% 100% |
| 2.0 | Haskell GHC #4 | 15.72 | 15.72 | 1,056 | 984 | 0% 0% 0% 100% |
| 2.0 | Lisp SBCL #2 | 15.72 | 15.73 | 4,628 | 906 | 0% 0% 0% 100% |
| 2.0 | Go | 15.74 | 15.75 | 1,744 | 411 | 0% 0% 0% 100% |
| 2.0 | Pascal Free Pascal | 15.76 | 15.77 | 8 | 423 | 0% 1% 0% 100% |
| 2.0 | C gcc #3 | 15.76 | 15.77 | 644 | 463 | 0% 0% 0% 100% |
| 2.1 | Java 7 #2 | 16.13 | 16.14 | 14,924 | 950 | 1% 0% 0% 100% |
| 2.1 | Scala | 16.52 | 16.53 | 18,484 | 404 | 0% 1% 0% 100% |
| 2.1 | Scala #2 | 16.77 | 16.77 | 18,764 | 720 | 1% 0% 1% 100% |
| 2.1 | Java 7 | 16.87 | 16.88 | 15,416 | 514 | 1% 0% 0% 100% |
| 2.3 | Lisp SBCL #3 | 17.99 | 18.00 | 4,640 | 883 | 0% 0% 0% 100% |
| 2.4 | OCaml #2 | 18.85 | 18.85 | 1,644 | 377 | 0% 0% 0% 100% |
| 2.4 | OCaml #3 | 17.97 | 18.91 | 3,252 | 938 | 0% 0% 0% 100% |
| 2.5 | Dart | 19.71 | 19.72 | 38,904 | 477 | 0% 0% 0% 100% |
| 2.6 | C gcc | 20.65 | 20.66 | 364 | 383 | 0% 0% 0% 100% |
| 2.6 | ATS | 20.68 | 20.69 | 608 | 1138 | 0% 0% 0% 100% |
| 2.6 | Go #2 | 20.75 | 20.76 | 1,256 | 668 | 1% 0% 0% 100% |
| 2.6 | Go #3 | 20.77 | 20.78 | 1,888 | 536 | 0% 0% 0% 100% |
| 2.8 | Ada 2005 GNAT | 21.76 | 21.77 | 1,232 | 710 | 0% 0% 0% 100% |
| 2.9 | Racket #3 | 22.88 | 22.89 | 18,540 | 627 | 0% 0% 0% 100% |
| 3.1 | Lisp SBCL | 24.70 | 24.71 | 5,712 | 625 | 0% 0% 0% 100% |
| 3.6 | Racket #2 | 28.42 | 28.44 | 17,132 | 532 | 0% 0% 0% 100% |
| 3.6 | C++ g++ | 28.47 | 28.48 | 872 | 452 | 0% 0% 0% 100% |
| 3.8 | C# Mono #2 | 29.49 | 29.49 | 14,108 | 1063 | 0% 0% 0% 100% |
| 3.9 | F# Mono #2 | 29.74 | 30.58 | 16,428 | 852 | 0% 0% 0% 97% |
| 3.9 | C# Mono | 30.98 | 30.99 | 13,872 | 459 | 0% 0% 0% 100% |
| 4.0 | Racket | 31.57 | 31.59 | 16,612 | 446 | 0% 0% 1% 100% |
| 5.7 | Haskell GHC #2 | 44.60 | 44.62 | 3,692 | 403 | 0% 0% 0% 100% |
| 5.7 | Erlang HiPE #2 | 45.12 | 45.13 | 10,164 | 747 | 0% 0% 0% 100% |
| 6.5 | Clojure #6 | 51.29 | 51.32 | 70,744 | 808 | 0% 0% 0% 100% |
| 6.6 | Clojure #7 | 51.75 | 51.79 | 66,164 | 762 | 0% 0% 0% 100% |
| 7.5 | Erlang HiPE | 59.14 | 59.16 | 13,652 | 507 | 0% 0% 0% 100% |
| 12 | Smalltalk VisualWorks #2 | 95.34 | 95.40 | 27,236 | 438 | 0% 0% 0% 100% |
| 28 | Perl | 216.22 | 216.31 | 3,640 | 333 | 0% 0% 0% 100% |
| 31 | Lua | 243.58 | 243.64 | 940 | 329 | 0% 0% 0% 100% |
| 41 | Ruby JRuby #4 | 5 min | 5 min | 595,740 | 326 | 0% 0% 0% 100% |
| 47 | Ruby JRuby #2 | 6 min | 6 min | 595,764 | 776 | 0% 0% 0% 100% |
| 51 | PHP #2 | 6 min | 6 min | 4,612 | 397 | 0% 0% 0% 100% |
| 52 | PHP #3 | 6 min | 6 min | 14,212 | 1193 | 0% 0% 1% 100% |
| 53 | Ruby JRuby | 6 min | 6 min | 595,616 | 292 | 0% 0% 0% 100% |
| 91 | Ruby 2.0 #3 | 11 min | 11 min | 25,120 | 828 | 0% 0% 0% 100% |
| 97 | Ruby 2.0 #4 | 12 min | 12 min | 5,536 | 326 | 0% 0% 0% 100% |
| 121 | Perl #2 | 15 min | 15 min | 2,784 | 343 | 0% 0% 0% 100% |
| 123 | Perl #3 | 16 min | 16 min | 7,320 | 846 | 0% 0% 0% 100% |
| 130 | Python 3 #6 | 16 min | 16 min | 5,208 | 328 | 0% 0% 0% 100% |
| 131 | Ruby 2.0 | 17 min | 17 min | 5,484 | 292 | 0% 0% 0% 100% |
| 138 | Python 3 #5 | 18 min | 18 min | 27,640 | 437 | 0% 0% 0% 100% |
| 146 | Python 3 #8 | 19 min | 19 min | 4,608 | 449 | 0% 0% 0% 100% |
| 180 | C CINT | 23 min | 23 min | 6,180 | 394 | 0% 0% 0% 100% |
| Scala #3 | Failed | 982 | ||||
| Scala #4 | Failed | 1006 | ||||
| "wrong" (different) algorithm / less comparable programs | ||||||
| 0.5 | C gcc #2 | 3.76 | 3.77 | 236,904 | 669 | |
| 1.1 | C++ g++ #7 | 8.30 | 8.31 | 924 | 1283 | |
| 1.5 | Python 3 #2 | 11.92 | 11.93 | 483,416 | 233 | |
| 2.1 | C++ g++ #2 | 16.33 | 16.34 | 924 | 1330 | |
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.