OQLQuery always goes to the database, because there is no easy way to tell if all rows interested visible to current transaction are already in cache.
For simply data object model, which has no relationship, it is expected to have similar timing. Thomas -----Original Message----- >From: Jackson, Scott M [mailto:[EMAIL PROTECTED]] >Sent: Wednesday, January 02, 2002 6:36 PM >To: [EMAIL PROTECTED] >Subject: [castor-dev] Does caching improve read performance? > >Hello , > >>From what I have been able to read, if castor does an oqlquery on an object >(thereby caching it from the database), the next time a similar read is done >should be from the cache, thereby significantly enhancing read performance. >I have benchmarked multiple SELECTs on the same object with both >cache-type="none" and cache-type="unlimited" (10 of each in succession) and >got identical timings. Watching the database logs, I saw the same selects >were performed for each and every query. It appears that no caching is >occurring. These 10 SELECTs are done in independent transactions (and with >new JDO's) and with a new DataBase object for every query as recommended to >avoid concurrency issues. The server of course is up for the all the >transactions. Am I doing it wrong or what am I missing? > > >Thank you, > >Scott Jackson > >p.s. I tried it with a single instantiation of jdo with the same result >(timing is identical -- a little faster for both sets though). > >----------------------------------------------------------- >If you wish to unsubscribe from this mailing, send mail to >[EMAIL PROTECTED] with a subject of: > unsubscribe castor-dev > ----------------------------------------------------------- If you wish to unsubscribe from this mailing, send mail to [EMAIL PROTECTED] with a subject of: unsubscribe castor-dev
