Hi everyone, I'm new here. I'm wondering if you guys can help...
I read a few other posts on this forum and there are quite a few things that are mentioned that confuse me. Can you guys tell me what OLAP, OLTP, Data warehouses and Data mining are. I have done a lot of research on this and this is what I have come up with:
OLAP - On-line analytical processing - Help to query large databases in the region of a billion rows. Mainly for performance issues? If that is the case can't we use SQL to query a database using a powerful server that can take care of the degrade in performance?
Data mining - This looks for statistical trends in a databse right? but arn't OLAP and data mining the same thing?
OLTP - On-line transaction processing - Is this typical database that is connected to say for example a website that handles commerce transactions or other stuff (mainly database driven websites) and has real time data inserted. This forum for example has a DB backend, is that an example of an OLTP?
Data warehouse - keeps track of historical data used for analysis by tools such as OLAP and data mining?
I hope tha above make some sense and would really appreciate if someone can clear all that up, as I have other questions that I want to ask but don't want to seem over-ignorant just yet.
P.s. I just noticed this is in the wrong place. If a moderator or admin can remove it to the Database design and Concepts forum I would appreciate it much.
OLAP and OLTP are kind of like the "Full" and "Empty" on your fuel gauge... They aren't really meaningful in a pure sense, but they give you a better feel for what is going on. I've only rarely seen applications that were pure OLTP, and never seen one that was pure OLAP at least as I think of the terms. Most applications fall somewhere in the spectrum between OLTP (which is the detail oriented, keep the wheels on the bus kind of stuff) and the OLAP (which is the "airplane view" looking down from high above stuff). Some folks in a business are nearly pure OLTP, they're involved in the day-to-day operations of the business itself. Other folks are almost completely OLAP, thinking where do we need to be in 6 months or 6 years (or in Japan in 6 centuries).
Data Mining is the process of getting a few snippets of information out of a mountain of data. Having lots of data is a good thing, but it doesn't directly allow action. Information is distilled data, and it is one of the most powerful "change agents" there is for any organization.
Also, with OLTP, think "Read-Write" for the user, whereas with OLAP, think "Read-Only"
OLTP often requires that data be entered/updated and read on a routine basis. With OLAP, the data is updated periodically, and then read many times.
This also means that indexing would be handled differently. Indexes speed up data retrieval, but slow down data insertion. Whenever you write a record, indexes must also be updated. So, with OLTP, you need to be judicial in how you apply indexing. With OLAP, since it is written infrequently, by standardized data migration routines, you tend to have more indexing, since this seeds up data retrieval. (and, you can disable indexing when you write all the records, then rebuild the indexes after the data has been written.)
Last edited by loquin; 01-26-06 at 18:32.
"Lisa, in this house, we obey the laws of thermodynamics!" - Homer Simpson
"I have my standards. They may be low, but I have them!" - Bette Middler
"It's a book about a Spanish guy named Manual. You should read it." - Dilbert