>- see footer for list info -<
Terry,
> At least, if I understand it right, with memcached we could cache a
> query (each one having an individual key) and if a related table is
> changed with an SQL update/delete/insert in our code, we can designate
> an immediate dropping of the named/related cached query. That way it is
> ready to be recreated the next time it is called for.
Ah, that reminds me of a quote by Phil Karson[1]:
"There are only two hard things in Computer Science: cache
invalidation and naming things."
We use memcached with CF8 heavily on White Label Dating[2] to cache
both data structures (e.g. complete member profiles) and pre-rendered
HTML pages (and sections of pages).
We don't use it directly for storing CF query data because the
serialisation/deserialisation is very heavy for those objects, and we
find it's much more efficient just to store the data we actually need
without any extra columns, query metadata etc.
We use a customised version of Shayne Sweeney's CF client[3] as a base
with a couple of façade components sitting on top to provide custom
handling of things like initialising the client, setting cache
timeouts, using counters, batching requests together nicely etc.
If you've got any specific questions about implementing memcached,
fire them over and I'll see what I can do to help.
Tim.
http://tim.bla.ir/
[1] US programmer from Xerox/DEC/Netscape:
http://www.nndb.com/people/400/000031307/
[2] http://www.whitelabeldating.com/
[3] http://memcached.riaforge.org/
_______________________________________________
For details on ALL mailing lists and for joining or leaving lists, go to
http://list.cfdeveloper.co.uk/mailman/listinfo
--
CFDeveloper Sponsors:-
>- cfdeveloper Hosting provided by www.cfmxhosting.co.uk -<
>- Lists hosted by www.Gradwell.com -<
>- CFdeveloper is run by Russ Michaels, feel free to volunteer your help -<
