Re: Raw Devices: Increased Performance?

From: Willy Klotz <willyk_at_t-online.de>
Date: 1996/06/25
Message-ID: <4qnu4l$vn8_at_news00.btx.dtag.de>#1/1


answers_at_ix.netcom.com(Steve Long) wrote:

>Jo,
 

>I strongly recommend you DO NOT use raw devices. Although you will see
>a minimal performance gain ( < 10%), you will create a significant
>increase in administrative overhead managing back-up and recovery as
>well as considerably more time performing a recovery in the event of a
>device failure. Economically speaking, it is cheaper to buy more
>memory and processors to gain considerably more performance than to
>spend the additional time on administrative tasks (personnel time) when
>raw devices are used for minimal performance gains.

This is an endless discussion.....

What I cannot understand, however, is the always brought up argument about the "administrative overhead".

At our site, we constantly use raw devices and we had a performance gain of 20 %; this is because our file-cache is used heavily by small reports, sometimes dozends per minute.

Back to the topic, where is the administrative overhead ? When you need a new datafile, then you have to define a file in a filesystem or, using raw devices, a slice of disk; in both scenarios you have to make sure that you have enough free space on your system.

The same goes, IMHO, in a recovery scenario - or what did I miss here ?

Regarding backup, it is simply a question of your backup procedure - if you use cpio or tar, you surely are out of luck...

>Steve
>804-262-6332

Willy Klotz


Willys Mail     FidoNet   2:2474/117  2:2474/118  
                Mailbox: analog 06297 910104
                         ISDN   06297 910105
                Internet: 0629791010_at_t-online.de                
                ->   No Request from 06.00 to 08.00 <-
======================================================================
Received on Tue Jun 25 1996 - 00:00:00 CEST

Original text of this message