Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: 3GB RAM usage by Oracle

Re: 3GB RAM usage by Oracle

From: Daniel Morgan <damorgan_at_x.washington.edu>
Date: Sat, 28 Aug 2004 09:30:33 -0700
Message-ID: <1093710688.869270@yasure>


Don Burleson wrote:

> "Richard Foote"
>
>

>>No, there's a third solution that's actually used by most shops.
>>
>>Have a DBA who actually knows *how* to tune a database and allocate memory
>>*appropriately*. Requiring AWE is rarely required and suggesting that memory
>>should simply be used because it's there, especially with the *significant
>>overheads* that AWE entails is poor advice even by your standards.

>
>
> Believe it or not, most USA shops already have tuned their SQL and
> their high physical I/O is due to not enough data cache. Things must
> be different where you live.
>
> Besides, it's now easy with v$db_cache_advice and 10g AMM to see the
> marginal benefit of adding RAM. Are you suggesting that AMM
> recommendations are nonsense?
>
> It's not uncommon to see working sets of frequently-referenced data of
> for than 30-gig for a large database. AWE is a great techniques for
> 32-bit Windows databases and I do it for dozens of databases every
> year, ALWAYS with great results.
>
> Also, you ignore the economic reality of database tuning. Time and
> time again, it's too costly (in both time and money) for a shop to
> tune their SQL. It's also an admission that the IT manager hired
> goofballs (usually a non-US provider) to write the original code.
> Like I said, I have no problem throwing hardware at crappy code when
> the client doesn't want to tune it.

The truth of the matter is that the type of tuning you and Richard are referring to is rarely going to get you anything significant compared with going to the low hanging fruit.

Most databases have performance problems for very simple reasons such as terribly written PL/SQL, failure to use bulk collection, failure to build appropriate indexes, failure to gather appropriate statistics.

Not that tuning memory configurations isn't a consideration but long after dealing with the basics. And almost always it is the basics that have been ignored.

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu
(replace 'x' with 'u' to respond)
Received on Sat Aug 28 2004 - 11:30:33 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US