On Mon, 21 Jun 2004 15:15:16 +1000, Pete Sharman <peter.sharman@xxxxxxxxxx> wrote: > > You just don't. Different block size tablespaces are there for transportab= > le tablespaces only. Absolutely zero benefit outside of that. > > One day soon I hope this message spreads a bit more broadly. Maybe we shou= > ld put it in a FAQ somewhere. :( <Red Rag to Bull> Absolutely zero? As in *never* useful outside of that? Possibilities that spring to mind. LOBS/Long/varchar2(4000) cols *might* benefit from larger block sizes, especially if the 'large' objects are actually quite small (say 8k in size). I would *guess* that small and very frequently updated LOBS would be good candidates - especially if PCTVERSION is set insanely. seperating these sorts of objects usage of the buffer cache out from that of 'ordinary' usage might be beneficial/more efficient (though I guess the recycle pool might be a good alternative). I don't like *never*. </Red Rag> <completely lost it> then of course we could put index and data segments in different block sized tablespaces to reap huge performance benefits (say 500%) in the same way as seperating them physically is good, seperating them in memory would be good too </completely lost it> Cheers -- Niall Litchfield Oracle DBA http://www.niall.litchfield.dial.pipex.com ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------