Taking the speed limiter off of VUGEN

4 views
Skip to first unread message

James Pulley

unread,
May 18, 2011, 2:47:28 PM5/18/11
to lr-loa...@googlegroups.com, LoadR...@yahoogroups.com

People ask me quite often Why is VUGEN so much slower than <fill in the blank>?  when what they really mean is How can I speed it up?

Well,   I have a set of observations which may help you.     Over the past week or so I have been working with some scripts which have a BASE64 encode and decode set of algorithms included.  Like many of you I sat there and waited and waited and waited for the encode or decode to finish.  I went into the bag of tricks to see what I could find and here are my findings.    Note, all times are on a Windows XP Pro box with a 3GHZ single core Intel processor, 1GB of RAM and 5GB of free disk space using a constant 64k file size, LR 9.x.  Benchmark time is for BASE64 ENCODE only.    Your mileage may vary

Condition                               Time (seconds)

=================               ===============

w/animated run                    144.1102

w/o animated run                  59.9814

        w/extended attribute

                -sensitivity 0            0.0449

So, for those of you looking to add a bit of pep to the ole VUGEN development, try the flag for sensitivity in the extended attributes.  You are probably thinking where the heck did this sensitivity flag come from?   Well, there are two hinted references.  The first is in LRUN.H.  The second is on the help dialog which comes up when you type MDRV with no options from the command line, assuming that you are in the <loadrunner home>\bin directory or this directory is in your path.

Be curious, you never know what kind of improvements in efficiency you can find lurking in the corners of LoadRunner.

James Pulley, http://www.loadrunnerbythehour.com/PricingMatrix

André Luyer

unread,
May 24, 2011, 3:44:33 PM5/24/11
to LoadRunner
Another hint: replace "extended attributes" by "additional attribute"
in James's text.
Changing LR_SENSITIVITY_HIGH did not work in my case (LR 9.52),
suggested somewhere else on the net.

It runs 300 times faster in my case (generating a large data file),
but breakpoints and step by step executing are no longer available
(who cares...). And 'animated run' setting was no longer relevant.
Biggest plus: it runs exactly the same way as on a load generator.

Anyway, good to know how to release the hand brake ;-), thanks James
for the tip.

Regards,
André

On 18 mei, 20:47, "James Pulley" <loadrunner-li...@jamespulley.com>
wrote:
> People ask me quite often "Why is VUGEN so much slower than <fill in the
> blank>?"  when what they really mean is "How can I speed it up?"
>
> Well,   I have a set of observations which may help you.     Over the past
> week or so I have been working with some scripts which have a BASE64 encode
> and decode set of algorithms included.  Like many of you I sat there and
> waited and waited and waited for the encode or decode to finish.  I went
> into the bag of tricks to see what I could find and here are my findings.
> Note, all times are on a Windows XP Pro box with a 3GHZ single core Intel
> processor, 1GB of RAM and 5GB of free disk space using a constant 64k file
> size, LR 9.x.  Benchmark time is for BASE64 ENCODE only.    Your mileage may
> vary
>
> Condition                               Time (seconds)
> =================               ===============
> w/animated run                    144.1102
> w/o animated run                  59.9814
>         w/extended attribute
>                 -sensitivity 0            0.0449
>
> So, for those of you looking to add a bit of pep to the ole VUGEN
> development, try the flag for sensitivity in the extended attributes.  You
> are probably thinking "where the heck did this 'sensitivity' flag come
> from?"   Well, there are two hinted references.  The first is in LRUN.H.
> The second is on the help dialog which comes up when you type MDRV with no
> options from the command line, assuming that you are in the <loadrunner
> home>\bin directory or this directory is in your path.
>
> Be curious., you never know what kind of improvements in efficiency you can
Reply all
Reply to author
Forward
0 new messages