RE: RAC/ASM Help
- From: "Mark W. Farnham" <mwf@xxxxxxxx>
- To: <duncan.lawie@xxxxxxxxxxxxxxxxx>, <shivaswamykr@xxxxxxxxx>, <oracle-l@xxxxxxxxxxxxx>
- Date: Mon, 23 Oct 2006 17:26:20 -0400
Right you are. Wrongly stated on my part. As far as I know, only three
things change the layout and structure: 1) Rebalancing following a storgage
change like adding or removing a disk, 2) Opening a new database file, and
3) extending an existing database file.
ASM does not yet react to ?heat? except when the the ?heat? results in
extending the database file. Then when the file extends, instead of possibly
staying on a single spindle in contiguous space avoiding seeks for the
streaming batch case, the new blocks will be spread evenly across the group.
So ASM can in fact mess up planned sequential streaming, but I attributed it
to entirely the wrong reason (wishful thinking and a brain cramp on my
part).
Regards,
mwf
-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx]On
Behalf Of Lawie, Duncan
Sent: Monday, October 23, 2006 10:17 AM
To: 'mwf@xxxxxxxx'; shivaswamykr@xxxxxxxxx; oracle-l@xxxxxxxxxxxxx
Subject: RE: RAC/ASM Help
Shiva,
have a look at this EMC / Oracle Joint Paper for ASM on EMC
http://www.oracle.com/technology/products/database/asm/pdf/asm-on-emc-5_3.pd
f
Mark,
Do you have heavily i/o intensive batch jobs that have a potential to stream
(which may be interpreted by ASM as a hot spot when in fact it is maximum
throughput with minimized seek overhead)?
Is this meaningful? It was my understanding that ASM has no real
knowledge/understanding of "hot spots", and relies simply on spreading data
thinly to avoid them.
Cheers,
Duncan.
============================================================================
==
Please access the attached hyperlink for an important electronic
communications disclaimer:
http://www.credit-suisse.com/legal/en/disclaimer_email_ib.html
============================================================================
==
Other related posts:
