>
>
> Is there any way for me to cache large portions of web sites
> that don't change so that the proxy has to do very little
> retreiving of that web-site from the Internet. For Instance,
> google.com is often accessed by users on my network and I'm
> curious if there's any way of speeding up access to this site
> by only having the cache update the dynamic content while
> holding the static content for longer. How do I keep more of
> what is accessed saved in the cache? Has anyone faced this
> and come up with an efficient solution?
>
How will Squid know what's dynamic and what's static info ?
Basically it's the responsability of the remote webserver
to give freshness info about objects. Caches will obey
this info and act accordingly when deciding objects should
be updated or nor not.
Checkout :
http://www.mnot.net/cache_docs/
for technical info.
Google e.g. tells caches not to keep objects obtained
from it's server by setting Cache-Control to 'private'.
Check with :
http://www.ircache.net/cgi-bin/cacheability.py
M.
Received on Wed Mar 10 2004 - 23:45:29 MST
This archive was generated by hypermail pre-2.1.9 : Thu Apr 01 2004 - 12:00:02 MST