thread-ring benchmark N=50,000,000

Each chart bar shows how many times more Memory, one ↓ thread-ring 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
1.0OCaml #2 6 min5 min1,012350  27% 18% 19% 26%
2.1Go #5 14.6614.662,080405  0% 1% 1% 100%
5.0OCaml 11 min10 min5,040282  27% 26% 27% 27%
5.3C gcc #2 6 min5 min5,404575  21% 21% 25% 25%
5.4OCaml #3 6 min5 min5,488296  23% 22% 21% 23%
6.5C++ g++ 7 min5 min6,560636  24% 24% 29% 28%
7.3C gcc 6 min5 min7,412487  21% 21% 24% 23%
7.4C++ g++ #2 6 min5 min7,460588  23% 23% 21% 20%
7.9Haskell GHC 11.729.467,948306  9% 9% 100% 9%
9.5C gcc #4 6 min5 min9,576761  21% 20% 26% 25%
9.6C gcc #3 163.98164.079,676916  100% 1% 1% 1%
9.7Python 3 #2 8 min7 min9,848288  29% 27% 21% 22%
11C++ g++ #4 8 min153.2110,720572  83% 83% 82% 82%
11C++ g++ #5 8 min149.1610,816652  83% 83% 81% 81%
20Ada 2005 GNAT #3 8 min6 min19,788727  24% 24% 24% 23%
20Ada 2005 GNAT #6 202.4151.8819,8521015  98% 98% 98% 98%
20Ada 2005 GNAT #2 10 min6 min20,076560  29% 29% 30% 30%
22Ada 2005 GNAT 13 min7 min22,572602  43% 42% 38% 37%
22Ada 2005 GNAT #4 7 min6 min22,712960  15% 15% 33% 32%
31Ruby #2 26 min19 min31,308215  29% 29% 30% 30%
31Ruby 8 min7 min31,404331  28% 22% 23% 28%
44F# Mono #3 16.5416.5344,368329  64% 35% 2% 1%
48C# Mono 12 min6 min48,224476  46% 44% 43% 42%
48F# Mono #2 13 min5 min48,848555  60% 61% 60% 60%
49C# Mono #2 11 min8 min49,852591  32% 31% 33% 32%
52Java  #7 8 min6 min52,740473  30% 30% 21% 21%
65Rust 9 min7 min65,604473  27% 25% 25% 27%
115Racket 110.38110.44116,232262  1% 0% 100% 1%
320Perl 37 min27 min324,236353  35% 34% 32% 32%
380Lisp SBCL #2 8 min6 min384,736571  36% 18% 18% 35%
385Java  #3 8 min6 min389,172530  24% 24% 28% 27%
394Lisp SBCL 6 min5 min398,796618  25% 25% 26% 25%
398Clojure 123.94110.49403,028348  32% 32% 24% 23%
399Clojure #2 121.56108.63403,420299  32% 32% 23% 22%
723Perl #3 12 min10 min732,004489  30% 25% 25% 29%
792Erlang HiPE 41.0541.05801,964273  97% 1% 3% 2%
793Ruby JRuby 11 min9 min802,808342  30% 27% 26% 26%
Erlang Failed273
Pascal Free Pascal Make Error523
Ruby JRuby #2 Failed228
Scala Failed296
"wrong" (different) algorithm / less comparable programs
5.5Python 3 #3 10.1610.175,608270
11C++ g++ #3 8.688.6910,964726
20Ada 2005 GNAT #5 0.510.4920,2961476
28F# Mono #4 1.861.8728,520267
202Java  #6 1.010.94204,164543
284Java  #2 4.974.88287,356693
286Java  #4 28.5526.85289,336894
289Java  #5 14.2813.08292,264432
missing benchmark programs
Dart No program
Fortran Intel No program
Hack No program
PHP No program

 thread-ring benchmark : Switch from thread to thread passing one token

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

More specifically:

diff program output N = 1000 with this output file to check your program is correct before contributing.

Each program should create and keep alive 503 pre-emptive threads, explicity or implicitly linked in a ring, and pass a token between one thread and the next thread at least N times.

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

Similar benchmarks are described in Performance Measurements of Threads in Java and Processes in Erlang, 1998; and A Benchmark Test for BCPL Style Coroutines, 2004. (Note: 'Benchmarks that may seem to be concurrent are often sequential. The estone benchmark, for instance, is entirely sequential. So is also the most common implementation of the "ring benchmark'; usually one process is active, while the others wait in a receive statement.') For some language implementations increasing the number of threads quickly results in Death by Concurrency.

Programs may use pre-emptive kernel threads or pre-emptive lightweight threads; but programs that use non pre-emptive threads (coroutines, cooperative threads) and any programs that use custom schedulers, will be listed as interesting alternative implementations. Briefly say what concurrency technique is used in the program header comment.

Revised BSD license

  Home   Conclusions   License   Play