python source benchmark

[Cedric Adjih]

seung-won hwang <shwang5 at students.uiuc.edu> wrote:

> Hello,
>
> I'm trying to benchmark python source with clock() function, but
> my code below returns negative value. I am not sure if my code is
> wrong or clock() is not suitable for getting CPU time... Any comment
> would help me a lot. Thanks.
>
> My code:
>   scl=time.clock()
> 	...
>   fcl=time.clock()
>   print "Execution Time: =", fcl-scl

> Reference:
> clock () 
> Return the current CPU time as a floating point number expressed in
> seconds. The precision, and in fact the very definition of the meaning of
> ``CPU time'', depends on that of the C function of the same name, but in
> any case, this is the function to use for benchmarking Python or timing
> algorithms

Well according to Linux documentation for clock(3):

  BUGS
    The C standard allows for arbitrary values at the start of
    the  program;  take  the  difference  between  the   value
    returned  from  a call to clock() at the start of the pro­
    gram and the end to get maximum portability.
    
    Note that the time can wrap around.   On a  32bit  system
    where  CLOCKS_PER_SEC  equals  1000000  this function will
    return the same value approximately every 72 minutes.

  Since CLOCKS_PER_SEC is actually defined to be 1000000 in
/usr/include/bits/time.h, on my Linux machine, you will have
a wrap-around after 72 minutes (or less) under the same circumstances.

  On Unix, there is a os.times() that probably suffers from the
same problem, and a module resource (resource.getrusage(...)), which
isn't compiled in on my machine.

  I usually use the /bin/time command, time.time() (on a lightly
loaded machine), or the profile module of Python for computing
the time spent in a program (as another person indicated here).

-- Cedric