click-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Bob Schellink <sab...@gmail.com>
Subject Re: Performance, velocity Renderable and Decorator
Date Thu, 15 Jul 2010 12:58:03 GMT
Hi Finn,

On 15/07/2010 22:31, Finn Bock wrote:
> 
> I wanted to see if the Renderable[1] interface in velocity 1.6 could
> be used to improved performance in click. And indeed it can!
> 
> I have made a quick and dirty hack that makes click-bench CLICK test
> perform more than 40% faster. From 567 to 839 request per second
> (average of 3 run after 1 discarded warm up run).


Quite impressive. It's difficult to know what is the real speedup of the direct rendering
vs the
Decorator change. I've eyed Decorator before because it creates a new String instead of using
the
buffer.

Would you be able to provide a breakdown of the speedup from the Decorator change vs the direct
rendering?

So something like:

before: 100 request per second
decorator change (but still using Strings): 120 requests per second
direct rendering: 140 requests per second


> I think the Decorator change can be made backward compatible by adding
> a fast Decorator2 interface and eventually deprecating the slower
> Decorator interface.
> 
> The dependency on velocity.jar is IMO acceptable for such a large
> performance gain.


Does this mean we won't support JSP or Freemarker?

Kind regards

Bob

Mime
View raw message