In a traditional Celery vs. Orange evaluation, ComputerWorld tries to sensationalize an analysis of Oracle 10g vs. SQL Server management expenses. It’s like evaluating a Nova to a Mercedes. They are both automobiles, but the resemblances end there.
So, what do we make of this “study”? It begins by properly observing that individual expenses consist of the most of IT price range, and that decreasing the variety costly DBA’s is an area of chance of the IT administrator to reduce expenses.
Now I ask you, how do you expect a DBA to respond when studying a short content that is concentrated on decreasing the DBA employees size? I know how supervisors think, and this content could wind-up in a different arms (envision the pointy-haired administrator in “Dilbert”).
Over the years, I’ve studied enough seasoned research to suspicious when research may be one-sided, so I did some verifying. The organization who did the research (Alinean Inc.), has a website on their website that is organised by Microsoft organization (Migration Calculator), and the web page says “Microsoft Associate Program”, showing that the research was done by a Microsoft organization partner. At the end of that web page, also observe “This website organised for Microsoft organization by Alinean, Inc.”
Managers will at some point start knowing some of those “research figures”. Some of them at least. You see, somebody states that one DBA can provide 9.9 Oracle databases/31.2 Microsoft organization data base , another one refers to 15.2 Oracle databases/65.4 Microsoft organization data base. Probably the numbers are not far from truth although I have seen that with some automated the numbers can develop better (100-150 data source per DBA) although I can’t take seriously research providing me numbers about customers reinforced by DBA. Sorry, what users? Web customers, data base users/schemas?
It is not up to the variety of database a DBA manages or the gb per DBAs or even business. It is up to the skill-sets and encounter of person doing the database organizations, it is up to atmosphere and the DBA group managing the performance. It is much about the set-up of the database and their complexness. Of course the automated associated with the DBA performance is an important aspect as well. But overview like the ones above and below are just slightly crazy. If someone tries to persuade you that it could help you’ll preserve a lot of cash by placing your database in “the Cloud”, just convert and run.
But wait around, there’s more!
Here we see a recommendation that Oracle data source are somehow “more work” than SQL Server data source, in this case, by an aspect of 3:1:
“Study members revealed that generally a database manager could handle more than 30 Microsoft SQL Server 2005 data source, while Oracle 10g implementations needed one DBA per 10 data source.”
I can’t talk for the common DBA, but I’ve individually handled over 50 Oracle data source, and with the appropriate automated in-place, a good DBA can handle many circumstances. Also, observe the classification of “large amounts of users” in the technique area of the study.
Oracle expert Mark Seeker (there are two well-known Mark Predators in Oracle), indicates that this wad-of-phlegm research is essentially useless because A linux systemunix is far more highly effective than Windows:
“I can’t believe people are actually compensated to come up with these figures. Allows see an oranges to oranges evaluation.
For example, put Oracle and MS SQL on the same components on Microsoft windows XP (yes, I’m going against my “Never Windoz” viewpoint, but last I examined, MS SQL was not available on Linux). “