On 30.11.12 17:18, Alex Rousskov wrote:
but categorizing a response before it enters the cache does make a lot of sense in many cases.
It does, except you don't necessarily know what categorisation is required at that point. Ideally it would be nice to categorise a request in respmod_postcache, and then store the categorisation results in the cache along side the object. That way if the object is later retrieved by a user who requires different categorisations, it can be analysed and those new results also attached to the object in the cache. However, that is going far beyond what squid is currently capable of.
Please note that the amount of overhead may actually _decrease_ with this solution: pre-cache + ACLs: analyze each miss + filter each access. post-cache: analyze and filter each access.
My current solution is to maintain an internal cache of analysis results within the ICAP server, so when a cached object is retrieved, only analysis that hasn't already been completed must be done; any analysis that was already done previously will be retrieved from the cache. This gives the best of both worlds, although as mentioned above the ideal would be to store the data alongside the object in the squid cache rather than maintaining two independent caches.
-- - Steve Hill Technical Director Opendium Limited http://www.opendium.com Direct contacts: Instant messager: xmpp:[email protected] Email: [email protected] Phone: sip:[email protected] Sales / enquiries contacts: Email: [email protected] Phone: +44-844-9791439 / sip:[email protected] Support contacts: Email: [email protected] Phone: +44-844-4844916 / sip:[email protected]
