Re: Oracle and Raid setup

From: Frank van Bortel <frank.van.bortel_at_gmail.com>
Date: Sat, 11 Jun 2005 10:26:13 +0200
Message-ID: <d8e6ti$3ue$1@news2.zwoll1.ov.home.nl>

xhoster_at_gmail.com wrote:
> Frank van Bortel <frank.van.bortel_at_gmail.com> wrote:
>

>>Noons wrote:
>>
>>>Cris Carampa apparently said,on my timestamp of 9/06/2005 4:41 PM:
>>>
>>>
>>>>But Raid 1 gives you only fault tolerance, no performance gains
>>>>because there is no striping. Or am I wrong?
>>>
>>>
>>>Absolutely right.  In fact, it is slower for writes...
>>>
>>
>>Why would a mirrored write be slower for writes? OK, as slow
>>as the slowest disk, but those differences can (should!) only
>>be marginal.

>
>
> If each write takes a random amount of time uniformly distributed
> between 0 and 1 (in whatever units of time would make sense), then the
> average wait for one write is 0.5, while the average wait for slower of two
> writes is 0.6667. That is a 33% penalty. I don't know if that qualifies
> as marginal or not. I don't know what a realistic actual distribution of
> write delays is, but I think uniform from 0 to 1 is not too outrageous of
> an assumption.
>
> Xho
>

I don't know what you are talking about, but it's not going to happen. Not in a mirrored setup, with "identical" disks - the quotes implying two disks, even bearing the same type code, will never be identical. There will be differences, but these will not be evenly distributed in a 0-100% range of the time.
Actual timing would be 8.0 millisecs vs 9.0 millisecs, so the OS would wait 9ms instead of 8 - so it could have been 11% faster.

But there is *no* penalty when you compare this mirrored setup to a single disk (of course, that disk is the 9ms disk :) ).

So again - were is the write penalty in RAID1 compared to a single disk?!?

-- 
Regards,
Frank van Bortel

Received on Sat Jun 11 2005 - 03:26:13 CDT