Data Warehousing Best Practices{1}


The Journal I decided to blog about this week is Titled, “Best Practices in Data Warehousing to Support Business Initiatives and Needs” by Jeff Lawyer and Shamsul Chowdhury. This journal follows a “diverse U. S. retailing company was experiencing the usual growing pains of the middle 1990’s” (Lawyer 2004). The company with faced with many technological and database problems including accessing data from their legacy systems, and the data contained little or no history. The company then decided in 1995 “that data warehousing could be used to release their data from its ‘data jailhouse'”(lawyer 2004). The journal speaks of two data warehousing styles, the “Bill Inmon Style” and the “Ralph Kimball Style”. The styles have many similarities  however where they differ is how the data is arranged in the data warehouse itself. The Inmon style used an “atomic-level, third-normal form (3NF) relational format in which to store extracted and transformed data”(Lawyer 2004), while the Kimball style uses a “multidimensional style ‘dimension and fact’ arrangement in which to store extracted and transformed data”(Lawyer 2004). In the case of the US retailing company, they chose to go the route of the Inmon style for various reasons.

Some of the best practices were shared by the journal and these best practices lead to the success of the companies data warehousing. One of the more basic one is to ensure there is ongoing business champion through out the entire process. On top of the ongoing business champion, the company must have industry experts on data warehousing for both validation and expertise deficiencies. The data warehouse must also be treated as an ongoing process and expansion should be done as needed. It is important, once a data warehousing architecture is chosen, to adhere to it from beginning to end. The image bellow show the architecture the company had chosen and stuck to throughout.

Having good metadata, and providing “education regarding the data warehouse and data mart structure and content, SQL coding techniques, access tools, data privacy, and any other requests you need to field from the business users” (Lawyer 2004) is also key.

I chose this article because it spoke about a real life example of a topic we have and will be discussing more in depth in class. It showed the struggles of the company and how those stuggles were lessened or even eliminated due to data warehousing. The company from the article continues to thrive, and the data warehouse has since grown to seven terabytes with two hundred tables and two thousand seven hundred columns. With information must more accessible and being must more robust, I do not foresee the company running into trouble in the near future when it comes to data.

The article spoke in good detail about the two types of data warehousing and compared them to the needs of the company. The company chose one style which was Inmon, but the article did a good job comparing and contrasting the two styles and how they would fit in certain scenarios. It made the point that one of the biggest dilemas in data warehousing is choosing the style itself. To make the correct decisions for the organization, those calling the shots should be completely informed on the types of data warehousing and how each would fit in the organization. This was a very informative article overall, despite being a bit technical when discussing the two types of data warehouses.

 

Lawyer, J.; Chowdhury, S.; , “Best practices in data warehousing to support business initiatives and needs,” System Sciences, 2004. Proceedings of the 37th Annual Hawaii International Conference on , vol., no., pp. 9 pp., 5-8 Jan. 2004
doi: 10.1109/HICSS.2004.1265515
URL: http://0-ieeexplore.ieee.org.opac.library.csupomona.edu/stamp/stamp.jsp?tp=&arnumber=1265515&isnumber=28293