RE: ANTs Data Server?

  • From: "Ruth Gramolini" <rgramolini@xxxxxxxxxxxxxxx>
  • To: <oracle-l@xxxxxxxxxxxxx>
  • Date: Thu, 2 Sep 2004 13:43:00 -0400

Our application (developed by an outside vendor) also uses "memory resident
tables" to hold this kind of static data. This feature was added to increase
the speed of the on-line part of the application.  Of course one of the lead
developers called them "memory residence tables" the entire time she worked
on the project, making me wince each time she said it.  And she was one of
the few native english speakers on the team.  They did work, however.


-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx]On Behalf Of Daniel W. Fink
Sent: Thursday, September 02, 2004 8:34 AM
To: oracle-l@xxxxxxxxxxxxx
Subject: Re: ANTs Data Server?

I have worked with web applications that would load up a local data
cache with 'static' information so that they did not have to access the
database to retrieve this type of data. Why should the app have to
perform the same query to return the same data, incurring network
overhead, database activity, etc. when it was totally redundant? In one
case, some of the data would only be changed during an application
release. Using my favorite example, every time a user is asked for their
state/province for an address, do you need to query the database to
build a list or validate their entry? By caching this static data
locally, you eliminate the need to access the database (via who knows
how many layers of network) every time.  If you process 100,000 user
address requests a day and only reload the data once every 24 hours, you
have reduced your app->database traffic by 99,999 queries each day.
Of course, this does not mean that the response time will be better.
Depending on the size of the data and how efficient a local search is,
it might be faster to access the database each time.

This is not the same as keeping all the data in the memory on the
database server.


ryan gaffuri wrote:

>has anyone worked with middle tier caching of read only data? About six
>months I went to a presentation by a non-database performance analyst(he
>analyzed the applications) and he was adament about caching data in the
>middle to relieve unnecessary hits on the database.
>has anyone exploited this?
>----- Original Message -----
>From: "Cary Millsap" <cary.millsap@xxxxxxxxxx>
>To: <oracle-l@xxxxxxxxxxxxx>
>Sent: Wednesday, September 01, 2004 9:58 AM
>Subject: RE: ANTs Data Server?
>>I just don't get the whole "cache everything, and everything will be ok"
>>argument. The vast majority of the slow tasks I've seen in the past ten
>>years have all executed "in cache" to begin with!
>To unsubscribe send an email to:  oracle-l-request@xxxxxxxxxxxxx
>and put 'unsubscribe' in the subject line.
>Recent messages -

To unsubscribe - mailto:oracle-l-request@xxxxxxxxxxxxx&subject=unsubscribe
To read recent messages -

To unsubscribe - mailto:oracle-l-request@xxxxxxxxxxxxx&subject=unsubscribe 
To read recent messages -

Other related posts: