Re: 3GB RAM usage by Oracle

Noons wrote:

> "Howard J. Rogers" <hjr_at_dizwell.com> wrote in message
> news:<4132279b$0$7222$afc38c87_at_news.optusnet.com.au>...
>

>> able to distinguish fact from fantasy. Next, you'll be wanting the man
>> that put this together (http://www.dba-oracle.com/redneck.htm) to show
>> some taste and decency. Or go on a diet.

Nah. I suddenly twigged why Don's earlier posts about indexes were so hopelessly, technically, inadequate. It's because in Don's part of the world, knowing how indexes actually behave is sheer luxury. You don't *need* to know such arcane details if all you're going to do is buy extra hardware.

It probably explains Mike Ault's demonstrated inadequacies on the issue, too.

Who gives a stuff about knowing anything about that sort of thing if you're gonna cream $10,000 for doing sod-all and recommend the obvious anyway?

It explains a lot, actually.

And it's not such a far-fetched tuning technique. Spend 10 minutes working out whether its memory, I/O or a CPU bottleneck, and then recommend purchasing excessive RAM, disks or CPUs, respectively, to fix it. As Don says, and I don't think we can dispute the matter, so long as you address the right bottleneck in your hardware purchasing, the thing will go faster after your investment than before. Probably.

That it equates to buying a fat man, who wouldn't know a diet if it bit him on his expansive behind, a Ferrari in which to zoom up and down the seaboard to no great effect... well, it doesn't matter. It's a Ferrari, after all.

Don's actually onto something here. I give you cheap shit versus expensive 'doing it properly'... well, people are going to go with Don every time.

He can even admit he brings no more talent than a two-day DBA to the party.... and he will still have clients, because they prefer cheap shit to decent work.

Extraordinary.

You learn a new trick every day. I merrily concede defeat. The prolific author has anyone who cares about doing a quality job beaten. 'Cos he can do a shonky job cheaper.

HJR Received on Mon Aug 30 2004 - 06:14:19 CDT