"Oracle's [multi-core] pricing ...simple and flexible"

The way I figured it was wrong....

"Now, while executives say that a new policy will account for each processor on multicore chips as a fraction -- 0.75 -- of a processor, there is a hitch.

"The new pricing will round up to the nearest whole number when the total number of processors housed on multicore chips totals between integers. Thus, a license for a two-core processor would charge for 0.75 x 2, which is 1.5 processors. In this instance, the enterprise's license would charge for the next highest integer, which is 2."

• Forwarded message ---------- From: Paul Baumgartel <paul.baumgartel_at_gmail.com> Date: Jul 15, 2005 2:09 PM Subject: Re: "Oracle's [multi-core] pricing ...simple and flexible" To: oracle-l_at_freelists.org

The way I figured it (based on 64-bit dual-core Opteron, which is supposed to give 50% better performance than single-core), the license cost is now commensurate with performance.

1 dual-core CPU = 1.5 CPU performance

1 dual-core CPU = 1.5 CPU license cost (2 x .75 CPU)

--
Paul Baumgartel
paul.baumgartel_at_aya.yale.edu



On 7/15/05, William B Ferguson <wbfergus_at_usgs.gov> wrote:


>

> Hmmm. Using the example below then, 12 cores equals the same 'processor'

> license as 11 cores?

>

> It's probably another set of licensing, but does this mean if we went with a

> 12 machine RAC, we'd only get counted as having 9 'cores', or would we

> actually then get counted for 12 'processors'?

>

>

> ------------------------------------------------------------

> Bill Ferguson

> U.S. Geological Survey - Minerals Information Team

> PO Box 25046, MS-750

> Denver, Colorado 80225

> Voice (303)236-8747 ext. 321 Fax (303)236-4208

>



-- 
Paul Baumgartel
paul.baumgartel_at_aya.yale.edu
--
http://www.freelists.org/webpage/oracle-l