RE: RE: Stripe and Mirror Everything - The "S.A.M.E. Method"

what if you don't always go to the Sam's Club (bulk store) for boxes of 48 rolls of toilet paper?
what if you don't like buying a side of beef at a time? Why keep a huge cache of stuff that will most likely go stale?

A 1 MB OS block size for indexed lookups? I thought that the point of using an index was to use the ROWID to obtain THE block (8KB) that you're looking for, not 127 of its close friends.

It looked like marketing directed at someone that wanted throughput on serialized batch jobs.
Its probably really good for that - but it looked far from one size fits all.
They started with moving the Online-Redo logs off to a different partition. Sounds good.
I tend to think that the reason that they stopped there, was it wouldn't be too many more steps (more distinct partitions) and they're right back to OFA - marketing premise right down the tubes ...

just my uniformed opinion - but at least I read the article.

Paul

• Original Message -------- Subject: RE: Stripe and Mirror Everything - The "S.A.M.E. Method" Date: Thu, 18 Jan 2001 14:11:22 -0800 From: "Steve Orr" <sorr_at_arzoo.com> Reply-To: ORACLE-L_at_fatcity.com Organization: Fat City Network Services, San Diego, California To: Multiple recipients of list ORACLE-L <ORACLE-L_at_fatcity.com>

I like your humble opinion and thanks for sharing it. Now my appetite is whet for a night of fine dining.

-----Original Message-----
Sent: Thursday, January 18, 2001 11:29 AM To: Oracle-L; Oracledba; 'Steve Orr'

Humble opinion on:

I think this approach will have strong appeal to IT Managers/Directors. It takes what I like to call the McDonald's approach to the computer center. I can walk into any McDonalds in this great country and I know what to expect in terms of experience, food and cost. They are all more or less the same (S.A.M.E !)

Managers find this approach to systems attractive for a number of reasons not limited to the fact that it appears to meet technical goals and the support staff are easier to find and they become interchangeable and outsourceable.

The paper states that the traditional approach "offers maximum configuration flexibility and performance tuning" but this works against "sameness" principles because a specialist with in-depth understanding of oracle and the application is required.

I think what they are offering is an approach that is uniform and simple. It may not be technically optimal but it does include IT management issues in the equation.

Humble opinion off !

p.s. I enjoy a Big Mac and fries just like the next guy but I'm not sure I'd like to eat there every day !

> ----------
> From: Steve Orr[SMTP:sorr_at_arzoo.com]
> Sent: Friday, January 12, 2001 9:54 AM
> To: Oracle-L; Oracledba
> Subject: Stripe and Mirror Everything - The "S.A.M.E. Method"
>
> Here's the url for a white paper titled "Configuring the Oracle Database
> with VERITAS Software and EMC Storage for Optimal Scalability,
> Manageability, and Performance" on Technet:
Received on Thu Jan 18 2001 - 20:08:19 CST