cocoon-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From rufio <ru...@op.pl>
Subject Re: forced caching of volatile data
Date Mon, 11 Aug 2003 13:19:48 GMT
on Mon, 11 Aug 2003 02:45:29 +0000 "Steve Schwarz"
<saschwarz@hotmail.com> wrote:

> You might want to talk to your DBA about how your database handles
> caching. "Properly" tuned if all this SQL activity is really all
> selects (and all the same set of selects) the DBMS shouldn't even hit
> the disk - all the data will be in it's in memory caches. Of course
> query design dramatically impacts cacheablity too. You'll pay the data
> transfer and cocoon processing cost but it might be "trivial"
> depending on your queries/amount of data if it all stays in the dbms
> cache.
> 
> You may really want/need the transactional integrity provided by the
> dbms if you are combining queries (on possibly different caching
> intervals) in a client single response.
> 
> As Geoff implies the approach you mention could provide stale data for
> as long a 59.99999...secs depending on the frequency of
> insert/update/delete. Rigging up a load test with the real queries/db
> will  tell you if this is an issue...

I'm not talking about DB, I want cacheable SQLTransformer.
Such transformer with some atribute, like 
<map:transform type="sql">
 <map:parameter name="forcevalidfor" value="60"/>
would assume validity of cached result for 60 secs.

> 
> HTH
> Steve


Regards, Rufio

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org


Mime
View raw message