Well, by that comment I was referring to the pre-page sga setting only.
With regard to the hugepage settings, typically you notice substantially
less CPU usage, especially on VMs. in my experience, it was about a 5%
reduction of CPU usage on physical servers, and somewhere between 10-20%
reduction on virtual servers.
On Tue, Feb 11, 2020 at 10:20 AM Ram K <lambu999@xxxxxxxxx> wrote:
This was going to be my next question: " I haven’t noticed any performance
gain from setting it true" .
How do I measure the performance gain? WIth 32Gb and 4K page size we are
looking at 8.3M memory pages; with 32G and 2M page size, we have 16K pages.
Will the responses be faster esp, with cases where several pages need to be
scanned? The DB equivalent I can think of is full table or index scans.
I was also also looking up the lock_sga setting. I am left wondering why
we even have the choice to set LOCK_SGA to FALSE. Should it not be true all
the time? The manual
<https://docs.oracle.com/en/database/oracle/oracle-database/12.2/refrn/LOCK_SGA.html#GUID-83BBA5B8-0E01-4B10-8278-AE4EEE39328B>
says that "This parameter is ignored on platforms that do not support it"
On Mon, Feb 10, 2020 at 7:39 PM Andrew Kerber <andrew.kerber@xxxxxxxxx>
wrote:
It can slow down the startup, but usually not a lot. However, I haven’t
noticed any performance gain from setting it true since 10.2.
Sent from my iPad
On Feb 10, 2020, at 19:35, Mladen Gogala <gogala.mladen@xxxxxxxxx>wrote:
SGA,but it will still slow down the startup of your database.
Pre-paging SGA will slow down your DB considerably. 32 GB is a small
TRUE
On 2/10/20 6:43 PM, Ram K wrote:
pre_page_sga boolean
--
--
Mladen Gogala
Database Consultant
Tel: (347) 321-1217
--
//www.freelists.org/webpage/oracle-l
//www.freelists.org/webpage/oracle-l
--
Thanks,
Ram.
