Walt Pawley wrote:
At 9:59 AM +0200 8/22/08, Oliver Fromme wrote:

- The perl command you wrote above is pretty much a sed
  command anyway (except you incorrectly used non-portable
  regular expression syntax).  Why use perl to execute a
  sed command?

At the risk of beating this to death, I just happened to
stumble on a real world example of why one might want to use
Perl for sed-ly stuff. I wanted to pull off the accessor's
address from each line of an Apache access log file. So, I
figured after this discussion that sed was the way to go. Then
I got curious and did the following:

wump$ ls -l Desktop/klog
-rw-r--r--  1 wump  1001  52753322 22 Aug 16:37 Desktop/klog
wump$ time sed "s/ .*//" Desktop/klog > kadr1

real    0m10.800s
user    0m10.580s
sys     0m0.250s
wump$ time perl -pe 's/ .*//' Desktop/klog > kadr2

real    0m0.975s
user    0m0.700s
sys     0m0.270s
wump$ cmp kadr1 kadr2

Why disparity in execution speed? Beats me, but my G5's fans
started to take off running the sed command. I don't think the
Perl command took long enough to register thermally. Curious.

FWIW: I did this with an older version of Mac OS X, rather
FreeBSD so it could easily not show the same results if I moved
the log file to a FreeBSD box and did it there.

Careful now.  Have you accounted for the effect of the klog file
being cached in VM rather than having to be read afresh from disk?
It makes a very big difference in how fast it is processed.

In order to get meaningful data for this sort of test you should
do a dummy run or two of each command in fairly quick succession,
and then repeat your test runs a number of times and look at the
average and standard deviation of the execution times. You'll often
see "Student's T test" mentioned -- that's a statistical test for
assessing if results calculated from a limited number of samples
represent different underlying distributions.  It sounds horribly
complicated, but nowadays we have computers to do all the difficult
adding up and the result is just a number that tells you how well
your supposition (that command 'a' is faster than command 'b') is
supported by your results.  There's a neat little script somewhere
that will automate that, and even give you an ascii graph output,
but I cannot for the life of me remember what it's called. Sorry.



Dr Matthew J Seaman MA, D.Phil.                   7 Priory Courtyard
                                                 Flat 3
PGP: http://www.infracaninophile.co.uk/pgpkey     Ramsgate
                                                 Kent, CT11 9PW

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to