cpython vs pypy

cpython
mmitchell:~/projects/formation/iweb_koans$ python --version
Python 2.7.2

mmitchell:~/projects/formation/iweb_koans$ /usr/bin/python profiling.py 
Filename: profiling.py

Line #    Mem usage    Increment   Line Contents
================================================
     4                             @profile
     5                             def test():
     6     7.922 MB     0.000 MB       # Create the big lists in advance to avoid skewing the memory counts
     7    15.555 MB     7.633 MB       seq1 = [None] * 10**6 # Big list of references to None
     8    16.320 MB     0.766 MB       seq2 = seq1[::10]
     9                             
    10                                 # Create and reference a lot of smaller lists
    11   141.488 MB   125.168 MB       seq1[:] = [[] for x in range(10**6)] # References all the new lists
    12   143.020 MB     1.531 MB       seq2[:] = seq1[::10] # Grab a second reference to 10% of the new lists
    13                             
    14                                 # Memory fragmentation in action
    15   150.652 MB     7.633 MB       seq1[:] = [None] * 10**6 # 90% of the lists are no longer referenced here
    16    69.047 MB   -81.605 MB       seq2[:] = seq1[::10] # But memory freed only after last 10% are dropped


Filename: profiling.py

Line #    Mem usage    Increment   Line Contents
================================================
    18                             @profile
    19     7.918 MB     0.000 MB   def hello():
    20                                 test()


done


mmitchell:~/projects/formation/iweb_koans$ pypy profiling.py 
Filename: profiling.py

Line #    Mem usage    Increment   Line Contents
================================================
     4                             @profile
     5                             def test():
     6    26.824 MB     0.000 MB       # Create the big lists in advance to avoid skewing the memory counts
     7    35.410 MB     8.586 MB       seq1 = [None] * 10**6 # Big list of references to None
     8    36.176 MB     0.766 MB       seq2 = seq1[::10]
     9                             
    10                                 # Create and reference a lot of smaller lists
    11   182.875 MB   146.699 MB       seq1[:] = [[] for x in range(10**6)] # References all the new lists
    12   182.875 MB     0.000 MB       seq2[:] = seq1[::10] # Grab a second reference to 10% of the new lists
    13                             
    14                                 # Memory fragmentation in action
    15   191.461 MB     8.586 MB       seq1[:] = [None] * 10**6 # 90% of the lists are no longer referenced here
    16   191.461 MB     0.000 MB       seq2[:] = seq1[::10] # But memory freed only after last 10% are dropped


Filename: profiling.py

Line #    Mem usage    Increment   Line Contents
================================================
    18                             @profile
    19    26.793 MB     0.000 MB   def hello():
    20                                 test()
    21                                 print "wow"


done