Re: Memory issue with Oracle 10g database on Sun Server T5140 with Solaris 10

From: Bradd Piontek <piontekdd_at_gmail.com>
Date: Thu, 16 Oct 2008 16:10:11 -0500
Message-ID: <e9569ef30810161410q288b0ed7te3f406c09926c003@mail.gmail.com>


Both those servers are T1 chipsets. I believe they came out 3 years ago or. The reason for the pricing, from my research, was that there weren't many multi-core chips out at the time, and to give Sun a fair pricing schema, they gave the .25/core pricing on it. When the T2 chip came out, quad cores were prevalent. All my research shows the T2 to be .75/core

Bradd Piontek
  "Next to doing a good job yourself,

        the greatest joy is in having someone
        else do a first-class job under your
        direction."
  • William Feather

On Thu, Oct 16, 2008 at 4:02 PM, Dan Norris <dannorris_at_dannorris.com> wrote:

> Took a little digging, but the Sun server table is at
> http://www.oracle.com/corporate/contracts/library/sun_server_table.pdf
>
> The reduced core multiplier is also listed in the definitions on the
> Technology price list (
> http://www.oracle.com/corporate/pricing/pricelists.html) on page 9.
>
> One does have to wonder why Oracle considers these CPUs to be less
> productive or at least that's how I interpret their move to discount the
> CPUs.
>
> Dan
>
>
> Bradd Piontek wrote:
>
>> As far as I know, the T2 chipset gets no special licensing from Oracle (it
>> is .75 per core like any other non-Intel/AMD processor). The T1 chipset that
>> was found on the T2000 had the .25 per core.
>>
>> There is an article from Sun that talks about all the oracle related
>> things you may see regarding performance on the CoolThread chips.
>>
>> http://blogs.sun.com/glennf/resource/Optimizing_Oracle_CMT_v1.pdf
>>
>>

--
http://www.freelists.org/webpage/oracle-l
Received on Thu Oct 16 2008 - 16:10:11 CDT

Original text of this message