Re: 12c pluggable database shared SGA question
- From: Connor McDonald <mcdonald.connor@xxxxxxxxx>
- Date: Wed, 17 Sep 2014 12:57:08 +0800
Regarding "I am, however, skeptical that there is any problem getting
enough DRAM bolted to modern CPUs"
This is true, but there (sadly) is large difference between what the
hardware is capable of, what the incredible cheap cost is.... versus some
infrastructure manager stuck in the 90's quibbling over an extra 16G to
bring the server up to a "ridiculous" 48G....
sigh....
I know some clients who are "excited" at multitenant as a way of reducing
their total SGA and they remain excited no matter how many times you tell
them just to head down the local IT shop and ffs just buy more RAM.....
On Wed, Sep 17, 2014 at 9:42 AM, John McHugh <JOHN.MCHUGH@xxxxxxxxxx> wrote:
> ...snip...
> i'd look for a database where bg cpu is higher than fg cpu. which could
> conceivably happen if the fg cpu is very very low. consolidate 50 of those
> completely idle databases on a single server and suddenly you've got a case
> where pdb may reduce cpu usage by 50%.
> ...snip...
>
> At last. Jeremy, you are right. Given a consolidation density and
> workload, comparing non-CDB (in the parlance) v. Multitenant, non-CDB
> consolidation efficiencies quickly degrade for the reasons you cite. BG
> processes compete for CPU, starving FGs, negatively impacting performance.
> This is particularly painful in commit processing where we see a lot of
> redo wastage in the non-CDB consolidation.
>
> Please review this doc, if you have not seen it before. And I would be
> very interested in your evaluations.
>
>
> http://www.oracle.com/technetwork/database/multitenant/learn-more/oraclemultitenantt5-8-final-2185108.pdf
>
> ... a very interesting discussion.
>
> thx
> jpm
>
>
--
Connor McDonald
===========================
blog: connormcdonald.wordpress.com
web: http://www.oracledba.co.uk
"If you are not living on the edge, you are taking up too much room."
- Jayne Howard
Other related posts:
