Re: oracle EE pricing

From: Kellyn Pedersen <>
Date: Fri, 22 Jan 2010 07:17:31 -0800 (PST)
Message-ID: <>

I just gotta say how much I love these kind of threads!   Oracle licensing and the verbage they use to get the most money out of any company's Oracle environment while the folks purchasing said licenses are simultaneously working through the text to find a way to get everything for less is just like watching a really good rugby match... :)

Kellyn Pedersen
Multi-Platform DBA
I-Behavior Inc.
"Go away before I replace you with a very small and efficient shell script..."

  • On Thu, 1/21/10, Mark Brinsmead <> wrote:

From: Mark Brinsmead <> Subject: Re: oracle EE pricing
To: "Bradd Piontek" <> Cc:, "" <> Date: Thursday, January 21, 2010, 9:46 PM

exactly.  The "trick" here is to find a hex-core processor (or any core-count, for that matter) that is not implemented as a multi-chip-module.  Sadly, the license's language around this subject is disappointingly imprecise. 

It does not way, for example what kind of chips in the MCM are to be counted as "sockets", so a (fictitious) quad core processor implement with:

two dual-core processor dies, plus
4 SDRAM chips, plus
2 clock multiplierscould -- perhaps ligitimately -- be considered as eight "sockets" (2 + 4 +2 = 8, as I recall) under the terms of the OLSA, and would therefore be completely ineligible for standard edition licenses.  This despite the fact that a well educated and (reasonably) well informed customer would probably count only the "processor" chips, and consider this to be two sockets, and thus potentially eligible for either SE or SE-1.

Personally, I think Oracle Corp really intended the "two socket" interpretation here, not the "eight socket" interpretation.  This is how I would expect a reasonable person (with perhaps a post-graduate degree in computer architecture) to interpret these rules.

Sadly, though, "little" questions like this one are only likely to resolved by lawyers.  And judges.  And courts.  And lots and lots of money.

On Tue, Jan 19, 2010 at 9:48 AM, Bradd Piontek <> wrote:

I"ve been trying to get my head around this interpretation for a while. I'm not sure this is correct (although it could be). A socket can have multiple cores on them. Some of the newer models don't implement th em via Multi-chip-modules. I can see how to Hex-Core chips could be used for SE/SE-One. A core <> a socket. 

Bradd Piontek
  The Pythian Group

On Tue, Jan 19, 2010 at 10:27 AM, Allen, Brandon <> wrote:

Mark, are you sure it’s permitted to license SE-1 on a 12-core server?  I thought SE1 could only be licensed on a max of 2 cores according to this:  
From p.2:
“Oracle Standard Edition One may only be licensed on servers that have a maximum capacity of 2 sockets. “  
From p.3:
“When licensing Oracle programs with Standard Edition One or Standard Edition in the product name, a processor is counted equivalent to a socket”  
Maybe I’m misinterpreting it?

From: [] On Behalf Of Mark Brinsmead

Yeah, $10K to license Oracle SE-1 on a 12-core database server

(And you can build a pretty darned powerful database server on SE-1 these days!)


Privileged/Confidential Information may be contained in this message or attachments hereto. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of this company shall be understood as neither given nor endorsed by it.


-- Mark Brinsmead

  Senior DBA,
  The Pythian Group


-- Received on Fri Jan 22 2010 - 09:17:31 CST

Original text of this message