Re: Oracle out the door

From: Bill Ferguson <wbfergus_at_gmail.com>
Date: Wed, 30 Apr 2008 12:47:36 -0600
Message-ID: <4025610e0804301147p7cdbd92bx37470749a0fd0ce5@mail.gmail.com>


On Wed, Apr 30, 2008 at 11:29 AM, John Thompson <jhthomp_at_gmail.com> wrote:
> That's what has mgmt. so upset. You can't selectively drop CPU licenses.
> It's all or none.
>

I work for a government agency. Our parent Department (which consists of about a dozen agencies) negotiated a contract for all the agencies, covering EE, Spatial and a few other 'options', with support. For anybody wanting to use additional 'features', they can buy those licenses at a reduced price with support.

This saves all of the agencies mucho dollars/year, and gives most people the core functionality they say they want. It also a lot cheaper than SQL Server, which even though we have an Enterprise contract with Microsoft, still costs $14K per license.In probably all but a couple instances, that's far more than what the various Oracle licenses cost us.

Oracle wants to keep you as a customer. A phone call or two to the sales folks (or their managers) should get things straightened out at a reduced yearly cost. If the sales rep can't get you a better price, ask to talk to their manager, or the regional manager. Not sure what it's like now, but when I first started using Oracle back in the early 90's, the sales force either met their sales quotas or were fired. Not sure how they computed customers who left, but I imagine it was counted against their quotas. Oracle needs that yearly money coming in, almost as much as they need the new sales. If customers leave almost as fast as new ones join, that leaves them with less market clout and increases their competitors.

Has anybody in your organization looked yet at what 'options' are needed for which databases, which can be consolidated onto a single server or instance, etc? Without knowing anything at all about your 97 databases management wants to 'convert', I'd venture that you can easily reduce the number of servers/instances at least in half, probably to one third or one fourth. License a couple EE boxes for the few that need the full blown capabilities of EE, and run most of the rest on SE boxes. If management thinks SQL Server will handle what those 97 other databases are doing, those probably don't need EE.

And always remember to remind management to calculate the ROI on converting all of those databases to SQL Server. And as bad as Oracle support can sometimes be, it is far superior to what Microsft has. Some systems can probably be converted to MySql or Postgres without losing any functionality fairly simply, but SQL Server will require major re-writes of code and thought processes.

-- 
-- Bill Ferguson
--
http://www.freelists.org/webpage/oracle-l
Received on Wed Apr 30 2008 - 13:47:36 CDT

Original text of this message