Talk:Data mart

From Wikipedia, the free encyclopedia

 This article is within the scope of WikiProject Databases.
??? This article has not yet received a rating on the assessment scale.
High rated as high-importance on the assessment scale

tzeh 5.jan.2004 wrote


A data mart is not a (logical) part of a data warehose (as described in the entry "data warehouse").

A data mart (physical) extract of the data warehouse which is made for separate purpuses.

These purposes may be:

  • special structure: e.g. OLAP für multidimensional analysis
  • special content: e.g. reduced parts of a huge data warehouse analogue to an user view
  • performance: separate handling on separete processional units (computers)
  • security: opening data for more or other users


Contents

[edit] Clean Up?

Is there anyway to request that this articled be cleaned up? The differing opinions of this Imnon person splashed about on the page give the otherwise straight-forward article an appearance of having been tagged by graffiti artists.

~~Flora

The article, which is opnionated and not factual, needs to be rewritten.

Martyn R Jones

[edit] Biased discussion

I've marked this article as needing a Neutral Point Of View. I've just starting to read up on the subjects of Data Warehousing and Data Marts, so I'm not yet capable of fixing this article.

Meanwhile, read some articles by the "opponent": http://www.dbmsmag.com/artin303.html#A000024

Michel Jansen

sorry, but you are basically saying that this article is not neutral but that you have no idea why. If you can't correct it or know what to add, then remove the NPOV banner--38.98.68.18 14:11, 21 September 2006 (UTC)

[edit] Biased

This article is biased. I think the Inmon "school of thought" is too pretentious. Mention of his name and a link to http://en.wikipedia.org/wiki/Bill_Inmon are both definitely in order given the significance of his contributions to the field, but he is by no means the sole authority on the subject. Just as important are references to Dr. Ralph Kimball and his work.

The Misconceptions section needs references.

Under 'Dependent Data Mart' there is a sentence reading "[...] a data warehouse team is inept and unable to create a usable data warehouse." I argue that 'unable' alone would convey the same meaning and in fact be more applicable: ineptitude can certainly be a cause for inability but I don't think this article should speculate on that matter.

--Ralgara 17:38, 26 July 2006 (UTC)

Wrong! The importance of Kimball is in the field of dimensional modeling and SQL, not in data warehousing. —Preceding unsigned comment added by 83.34.66.146 (talk) 15:53, 9 September 2007 (UTC)

[edit] Data Mart Complexity

It appears to me that the article is trying to convey too much meaning.

The concept linking seperate Data Marts (DM) to provide the functionality of an Enterprise Data Warehouse (EDW) is a comparison of Centralised vs Federated warehousing architectures and seems much less relevant to a posting on DM.

The comment on the Bill Inmon bias is valid. His views are pivotal but need to be ballanced with other major contributors.

My apologies for not feeling sufficiently competent to offer a re-write and only this comment. It does need help.

--HooglBoogl 08:31, 25 September 2006 (UTC)

[edit] Merge

Propose merging this with Data warehouse article. Comments? SqlPac 14:44, 17 May 2007 (UTC)

Definitely NOT to be merged. As pointed out on top, data marts and data warehouses are distinct entities, not to be confused. The fact that the distinction is often unclear makes this article more important.Nshuks7 20:32, 1 November 2007 (UTC)