Re: Oracle on RAID - Thanks

2003-09-07 Thread Mogens Nørgaard 
Yes. RAID-3 is bit-level striping with a dedicated parity disk. Buy 33 
disks if you're on 32 bit systems, otherwise buy 65. Then you'll have 
fantastic performance as long as you have concurrency for the whole disk 
stripe set to 1.

RAID-4, as used in the IBM Shark systems, have chunk-level striping (I 
can't think of the right word now) and a dedicated parity disk to ensure 
that you have at least one disk with a lot of constant pressure on it.

Jared Still wrote:

  Of course, the Sun engineer is saying our new RAID 5 is just as fast as
RAID 10 because we stripe parity across all disks.
   

Uh huh.  Does he think this is new?

RAID 3 is the version that has a dedicated parity disk IIRC.

RAID 5 has always striped the parity across all disks.

Jared

On Fri, 2003-08-22 at 08:24, DENNIS WILLIAMS wrote:
 

To whomever posted the link to Gaja's paper Implementing RAID on Oracle -
thank you. I previously had printed and passed this article on to my system
administrator -- no reaction. But yesterday he was building our new Sun
system with all the latest features, and was so excited about the paper that
he asked for a second copy to share with his junior sys admin. He stopped
building the system to take a day to study Gaja's paper in detail.
  Of course, the Sun engineer is saying our new RAID 5 is just as fast as
RAID 10 because we stripe parity across all disks.
   Since this will be an OLTP system, I am planning on an 8K Oracle block
size, unless anyone has better information on how to pick a block size.
Apparently there are a lot of storage system parameters you can optimize,
but they are based on the Oracle block size.


Dennis Williams 
DBA, 80%OCP, 100% DBA 
Lifetouch, Inc. 
[EMAIL PROTECTED] 

--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: DENNIS WILLIAMS
 INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).
   



 

--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: =?ISO-8859-1?Q?Mogens_N=F8rgaard?=
 INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).

Re: Oracle on RAID - Thanks

2003-08-23 Thread Tanel Poder 
This is exceptionally good from Gaja's doc:

A disk cannot spin faster than its design specification, cannot service
data-transfers that exceed its I-O bandwidth and cannot support a zillion
concurrent I-O operations.  This is true regardless who your storage vendor
is.


I think I'll get a few cans of spray and gonna paint it to the local storage
vendors' walls after dark. Or at least to few salesguys doors who sincerely
tell you that you don't have to worry about IO or file placement *at all*
since we got lot's of cache and the storage box sorts everything out on
his own. (Hm... btw, should storage arrays be called he or she? ;)

Tanel.

- Original Message - 
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Friday, August 22, 2003 6:24 PM


 To whomever posted the link to Gaja's paper Implementing RAID on Oracle -
 thank you. I previously had printed and passed this article on to my
system
 administrator -- no reaction. But yesterday he was building our new Sun
 system with all the latest features, and was so excited about the paper
that
 he asked for a second copy to share with his junior sys admin. He stopped
 building the system to take a day to study Gaja's paper in detail.
Of course, the Sun engineer is saying our new RAID 5 is just as fast
as
 RAID 10 because we stripe parity across all disks.
 Since this will be an OLTP system, I am planning on an 8K Oracle block
 size, unless anyone has better information on how to pick a block size.
 Apparently there are a lot of storage system parameters you can optimize,
 but they are based on the Oracle block size.



 Dennis Williams
 DBA, 80%OCP, 100% DBA
 Lifetouch, Inc.
 [EMAIL PROTECTED]

 -- 
 Please see the official ORACLE-L FAQ: http://www.orafaq.net
 -- 
 Author: DENNIS WILLIAMS
   INET: [EMAIL PROTECTED]

 Fat City Network Services-- 858-538-5051 http://www.fatcity.com
 San Diego, California-- Mailing list and web hosting services
 -
 To REMOVE yourself from this mailing list, send an E-Mail message
 to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
 the message BODY, include a line containing: UNSUB ORACLE-L
 (or the name of mailing list you want to be removed from).  You may
 also send the HELP command for other information (like subscribing).



-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.net
-- 
Author: Tanel Poder
  INET: [EMAIL PROTECTED]

Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).