by Polun L
This article, “Data Mart vs Data Warehouse”, by Justin Swanhart, talks about his experience of data mart and data warehouse and then point out the advantage of each tool. First, in his point of view, he mentions that data warehouse is actually a relational database schema that stores data from an operational system. Also, he adds, “the goal of data warehousing is to collect and make a historical record of the information from another system”. The data such as web application, will be stored into the warehouse organizationally. Another advantage of using data warehousing is that it is effectively for people to analysis data from the past and then predict the outcome in future. However, this type of warehouse never gets updated automatically because the database is for inserting and storing a large amount of data only. On the other hand, data mart tends to be updated frequently and was popularized by Ralph Kimball. According to the article, the author says, “The goal of this approach is usually multi-dimensional analysis as it is very hard to create a dimensional model from a highly normlized database schema”. what it means is that when data comes in, it is very hard to predict the model that is to be built.
What I have learned from this article is the data marts. It contains a star schema which consists of a central table and dimension tables such as lists of customers or products. Star schema can denormalize data partially, so tables in a data mart can be updated easily.
Justin, S. (2010, JULY 15). Data mart of data warehouse? [Web log message]. Retrieved from http://www.mysqlperformanceblog.com/2010/07/15/data-mart-or-data-warehouse/