vs

 1 : Are the PyPy programs faster? At a glance.

Each chart bar shows, for one unidentified benchmark, how much the fastest PyPy program used compared to the fastest Python 2 program.

(Memory use is only compared for tasks that require memory to be allocated.)


These are not the only compilers and interpreters. These are not the only programs that could be written. These are not the only tasks that could be solved. These are just 10 tiny examples.

 2 : Are the PyPy programs faster? Approximately.

Each table row shows, for one named benchmark, how much the fastest PyPy program used compared to the fastest Python 2 program.

(Memory use is only compared for tasks that require memory to be allocated.)

 PyPy used what fraction? used how many times more? 
Benchmark Time Memory Code
 n-body1/7 ±
 fannkuch-redux1/5 ±
 richards1/5 ±
 fib501/4 ±
 thread-ring1/4 ±
 jsonbench1/3 ±
 meteor-contest1/3 ±
 fasta1/3 ±
 fibonacci1/2 ±
 pystone1/2 ±
 iobench1/2 ±
 binary-trees±±
 chameneos-redux± ±
 k-nucleotide?±
 mandelbrot ±
 spectral-norm?
 templates ±
 regex-dna?±
 reverse-complement?±
 pidigits55×?±
 PyPy used what fraction? used how many times more? 
Time-used  |-  |---  25% median  75%  ---|  -|
(Elapsed secs)1/71/71/41/255×

± read the measurements and then read the program source code.

 3 : Are the PyPy programs faster? Measurements.

These are not the only tasks that could be solved. These are just 10 tiny examples. These are not the only compilers and interpreters. These are not the only programs that could be written.

For each named benchmark, measurements of the fastest PyPy program are shown for comparison against measurements of the fastest Python 2 program.

Program Source Code CPU secs Elapsed secs Memory KB Code B ≈ CPU Load
 n-body 
PyPy6.906.9072,4921337  100% 0% 0% 0%
Python 250.8250.826,7561337  100% 0% 0% 0%
 fannkuch-redux 
PyPy2.820.89285,6001009  78% 75% 97% 72%
Python 215.914.1342,7081009  95% 97% 96% 98%
 richards 
PyPy0.250.251,6802423  100% 4% 12% 0%
Python 21.171.177,1322423  1% 1% 100% 0%
 fib50 
PyPy0.650.6577,608136  0% 100% 0% 2%
Python 22.842.8438,432136  0% 0% 100% 0%
 thread-ring 
PyPy0.150.15?407  0% 0% 100% 0%
Python 20.550.557,072407  2% 100% 0% 2%
 jsonbench 
PyPy1.261.2679,572322  1% 2% 2% 100%
Python 24.364.3716,424322  0% 0% 100% 0%
 meteor-contest 
PyPy0.860.8682,2321579  1% 0% 100% 0%
Python 22.652.656,8681485  0% 84% 0% 16%
 fasta 
PyPy2.382.3977,848900  0% 0% 100% 0%
Python 26.466.4613,432900  0% 75% 0% 25%
 fibonacci 
PyPy0.280.281,636181  100% 0% 0% 0%
Python 20.660.677,128181  0% 0% 0% 100%
 pystone 
PyPy0.110.11?2301  0% 0% 0% 100%
Python 20.230.238442301  0% 100% 4% 4%
 iobench 
PyPy12.6912.6983,968370  0% 100% 0% 0%
Python 225.2125.216,804370  0% 79% 0% 21%
 binary-trees 
PyPy2.191.01322,228743  78% 70% 61% 58%
Python 23.771.1454,152743  82% 81% 82% 90%
 chameneos-redux 
PyPy103.6456.0078,3121192  34% 34% 37% 39%
Python 2113.1162.6911,0761192  34% 30% 32% 33%
 k-nucleotide 
PyPy0.110.12?593  100% 0% 0% 0%
Python 20.080.08?593  0% 11% 100% 0%
 mandelbrot 
PyPy130.0733.95126,2883832  96% 98% 98% 98%
Python 274.8719.17173,5243832  97% 98% 99% 97%
 spectral-norm 
PyPy0.150.15?594  100% 0% 0% 0%
Python 20.070.07?394  0% 0% 86% 0%
 templates 
PyPy2.662.6792,116307  0% 100% 0% 0%
Python 21.051.0516,332307  0% 0% 0% 100%
 regex-dna 
PyPy0.140.14?501  100% 0% 0% 0%
Python 20.030.03?501  100% 0% 0% 0%
 reverse-complement 
PyPy0.080.08?455  100% 0% 0% 0%
Python 20.010.01?455  0% 100% 0% 0%
 pidigits 
PyPy0.660.6686,928380  3% 3% 100% 0%
Python 20.010.01?380  0% 0% 0% 100%
 fasta-redux 
PyPy0.280.281,6921115  0% 57% 29% 19%
No program
 binary-trees-redux
   No programs

 4 : Are there other PyPy programs for these benchmarks?

Remember - those are just the fastest PyPy and Python 2 programs measured on this OS/machine. Check if there are other implementations of these benchmark programs for PyPy.

Maybe one of those other PyPy programs is fastest on a different OS/machine.

 5 : Are there other faster programs for these benchmarks?

Remember - those are just the fastest PyPy and Python 2 programs measured on this OS/machine. Check if there are faster implementations of these benchmark programs for other programming languages.

Maybe one of those other programs is fastest on a different OS/machine.

 PyPy : the old PyPy 

 

Revised BSD license

  Home   Conclusions   License   Play