Re: Enable 32K Block in 8K Block DB

"Frank van Bortel" <fvanbortel@netscape.net> wrote in message news:c4otl4$3fg$1@news3.tilbu1.nb.home.nl... > Bottom line (still) is - don't use buffered IO; use direct or raw > (why do you think every benchmark of oracle still uses raw?!?) Really Oracle benchmarks all use raw?...As an Informix person why did I used to keep hearing "according to Oracle raw devices are not faster, in fact filesystems are quicker" FROM MY CUSTOMERS than? ^^^^^^^^^^^^^^^^^^^^^^ Used to really annoy me...so Oracle have come around to my way of thinking than? As it's faster? What a suprise... > > Now - how do I convince the SA to go raw on his HP machine with VX7100, > where he found a benchmark that proofs he was right to create the > filesystem with 8k blocks? Dunno. > > -- > > Regards, > Frank van Bortel >

"David Williams" <djw@smooth1.fsnet.co.uk> wrote in message news:c4stsg$h3$1@newsg1.svr.pol.co.uk... > > "Frank van Bortel" <fvanbortel@netscape.net> wrote in message > news:c4otl4$3fg$1@news3.tilbu1.nb.home.nl... > > Bottom line (still) is - don't use buffered IO; use direct or raw > > (why do you think every benchmark of oracle still uses raw?!?) > > Really Oracle benchmarks all use raw?...As an Informix person why did > I used to keep hearing > > "according to Oracle raw devices are not faster, in fact filesystems are > quicker" > > FROM MY CUSTOMERS than? > ^^^^^^^^^^^^^^^^^^^^^^ > > Used to really annoy me...so Oracle have come around to my way of > thinking than? > As it's faster? What a suprise... The discussion is not so much whether raw is faster or slower than a file system, on which the jury is still out. The issue here is: with a file system that does buffered I/O, am I free to choose any Oracle block size I like? (nope). Can I achieve performance gains by adopting multiple blocksizes within the one database? (nope). That's a completely different discussion than the one you seem to want to start. Regards HJR -- ------------------------------------------- Dizwell Informatics: http://www.dizwell.com -A mine of useful Oracle information- -Windows Laptop Rac- -Oracle Installations on Linux- ===========================================

[cutting] > > The discussion is not so much whether raw is faster or slower than a file > system, on which the jury is still out. I'd disagree, the jury is in, all things being equal raw is faster. But all things are not equal :-)) so the debate goes on > The issue here is: with a file > system that does buffered I/O, am I free to choose any Oracle block size I > like? (nope). Can I achieve performance gains by adopting multiple > blocksizes within the one database? (nope). > > That's a completely different discussion than the one you seem to want to > start. > > Regards > HJR > -- > ------------------------------------------- > Dizwell Informatics: http://www.dizwell.com > -A mine of useful Oracle information- > -Windows Laptop Rac- > -Oracle Installations on Linux- > =========================================== -- Paul Watson # Oninit Ltd # Growing old is mandatory Tel: +44 1436 672201 # Growing up is optional Fax: +44 1436 678693 # Mob: +44 7818 003457 # www.oninit.com #