Re: continuation...

From: DA Morgan <damorgan_at_psoug.org>
Date: Tue, 17 Jul 2007 23:12:53 -0700
Message-ID: <1184739174.377995@bubbleator.drizzle.com>

Bob Jones wrote:
>>> Not to mention mission-critical systems that require near 100% uptime.
>> No one is more sensitive to that than Amazon.com.
>>
>> Want to guess what they're running on their production systems?
>>
>> How about the line of business applications at every mobile phone company
>> in the region: AT&T, T-Mobile, AllTel, Verizon, ClearWire?
>>
>> How about the line of business applications at banks like Washington
>> Mutual; one of the largest in the US?
>>

> 
> And what percentage of 10g are these companies running? I hope you have done 
> a survey internally. I have worked for some large companies. Even the DBAs 
> don't know where all the databases are.
>

>> Why is it some people here are getting their knickers in a twist and
>> yet these large organizations, with serious 7x24x365 requirements, are
>> not dropping like flies?
>>
>> And no it is not that they are lucky. They aren't buying it, and they
>> are not upgrading to 10gR2 because they are being bullied by some guy
>> with a $200 haircut. Nor would their CTO's risk their careers if the
>> systems were failing.
>>

> 
> If there are business needs, nothing can really stop a company from 
> upgrading. Technologies are there to support businesses. Some may think the 
> other way around. Not just CTOs, many people would be risking theirs jobs, 
> if they don't carefully consider the business impacts.

I often, due to what I do, find myself talking with IT management at a sufficiently high level to know what they are doing. The driving need is compliance and CTOs and CFOs are very aware of what can put them into legal jeopardy.

>> When you look around and ships aren't falling off the edge of the earth
>> you conclude that your premise that the earth is flat is wrong. When
>> you look around and see the largest most important organizations on the
>> planet running their line-of-business applications on 10g you should
>> conclude that they too are not falling off the edge of the earth.
>>

> 
> Sure, every large company has 10g running somewhere, but it is not just as 
> simple as upgrading Oracle. There are issues like coordinating with users, 
> application testing, bug fixing, downtime scheduling, and so on. It can take 
> 6 months to upgrade just one application. Think about 50 applications.

I do. But, as I said, it has been 3 to 4 years. And think about organizations, Boeing for example, that think training 300 DBAs on the Grid Control reasonable so you can take a good guess at the number of applications. Still ... they are moving, apparently, far faster than most.

> We do not upgrade just to be on the lastest. There has to be sufficient > business reasons.

There is: Sarbanes-Oxley, PIPEDA, BASEL-II. That is what is driving it here. I can't imagine which part of violating laws with serious consequences you wouldn't consider "sufficient business reason."

Let me give you just one small example. Sitting on my desk is an article from eWeek titled "Intel in hot water over lost e-mails." Try this link to see the implications.

http://www.investors.com/editorial/IBDArticles.asp?artsec=17&artnum=2&issue=20070321

Read the following paragraph from the link very carefully:

"One example is a suit against Morgan Stanley (MS) by billionaire Ronald Perelman in 2003 in Florida, claiming the bank defrauded him during a stock swap. The case turned in part on Morgan Stanley's inability to find e-mails that Perelman's attorney had requested.

In May 2005, a jury handed down a $1.45 billion verdict against Morgan Stanley."

1,450,000,000 sufficient business reason!

If you don't keep your IT management in the dark about the dark-ages practices of your current software's capabilities you may find them very interested in moving to 10g, 11g, audit vault, and other contemporary technologies. Likely the only reason they aren't leading the charge is that no one has sat down with them and given them a realistic appraisal of risk vs reward.

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu (replace x with u to respond)
Puget Sound Oracle Users Group
www.psoug.org

Received on Wed Jul 18 2007 - 01:12:53 CDT