Paulo Gaspar wrote:
> Notice that my focus is just cache efficiency. And, unlike what I have been
> able to transmit, I am interested on significant improvements.
Than we share the same concern :)
[skipped a good example of the need for a sampling key for resource
families]
I think I got you idea in full detail now and I agree it's a very good
suggestion.
The idea is to have the cacheable component logic create both the
caching key for the handled request *AND* a (possibly different)
sampling key.
So, for example, asking for the resource:
cocoon://reviews/movies/harry-potter
might generate a caching key that hashes the entire URI, while might
generate a sampling key that hashes only the general part
cocoon://review/movies/
thus, every request in this category will be collected in the same
statistical data pool, thus converge sooner and talking less memory.
It's a great idea and, yes, I totally missed it in my proposal.
There are some details to think about but I think it's a great addition
to my model. Thanks for sharing it with us. :)
--
Stefano Mazzocchi One must still have chaos in oneself to be
able to give birth to a dancing star.
<[EMAIL PROTECTED]> Friedrich Nietzsche
--------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
