Just FYI, I tried out how the old Links 2.0 performs, if we replace just one part of it: the template parser.
I was curious, how much performance can I get using GT::Template over the old Template.pm?
Maybe several of you also asked this question from yourself, so I thought this info would helpful for you.
BTW: Don't blame me about how the tests was done! Different live tests would likely show different results! This test was based on my live site, without the goal to do a very accurate test. However IMO, it shows the power of GT::Template...
I changed the parser to GT::Template and executed the same build event using each parser.
You are curious for the results, eh?
Well, the results are what I expected. GT::Template is much-much faster, than Template.pm.
Here they are the numbers:
1) Test results after parsing 1 page:
Template.pm: Time spent template parsing 1 template: (0,015 s)
GT::Template: Time spent template parsing 1 template: (0,002 s)
That's 7.5x faster!
One page does not really shows the possibilities, so we should see the test results after parsing more than 1 hundred pages.
2) Test results after parsing more than 1 hundred pages:
Template.pm: Work time of all template parsing: (15,43 s)
GT::Template: Work time of all template parsing: (0,90 s)
That's more than 17x faster than old Template.pm!
That's all folks!
I hope you enjoyed!
Best regards,
Webmaster33
Paid Support from Webmaster33. Expert in Perl programming & Gossamer Threads applications. (click here for prices)
Webmaster33's products (upd.2004.09.26) | Private message | Contact me | Was my post helpful? Donate my help...
I was curious, how much performance can I get using GT::Template over the old Template.pm?
Maybe several of you also asked this question from yourself, so I thought this info would helpful for you.
BTW: Don't blame me about how the tests was done! Different live tests would likely show different results! This test was based on my live site, without the goal to do a very accurate test. However IMO, it shows the power of GT::Template...
I changed the parser to GT::Template and executed the same build event using each parser.
You are curious for the results, eh?
Well, the results are what I expected. GT::Template is much-much faster, than Template.pm.
Here they are the numbers:
1) Test results after parsing 1 page:
Template.pm: Time spent template parsing 1 template: (0,015 s)
GT::Template: Time spent template parsing 1 template: (0,002 s)
That's 7.5x faster!
One page does not really shows the possibilities, so we should see the test results after parsing more than 1 hundred pages.
2) Test results after parsing more than 1 hundred pages:
Template.pm: Work time of all template parsing: (15,43 s)
GT::Template: Work time of all template parsing: (0,90 s)
That's more than 17x faster than old Template.pm!
That's all folks!
I hope you enjoyed!
Best regards,
Webmaster33
Paid Support from Webmaster33. Expert in Perl programming & Gossamer Threads applications. (click here for prices)
Webmaster33's products (upd.2004.09.26) | Private message | Contact me | Was my post helpful? Donate my help...