User Tools

Site Tools


concepts:data-breakdown-structure

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
concepts:data-breakdown-structure [2017/04/21 19:26]
LJuraschek created
concepts:data-breakdown-structure [2020/08/20 13:27] (current)
Line 1: Line 1:
-Data Breakdown Structure+====== Data Breakdown Structure ====== 
 + 
 + 
 +Data Breakdown Structure is a concept coined by Dan Linstedt and Michael Olschimke in their book //"Building a Scalable Data Warehouse with Data Vault 2.0"//, which defines the term as follows:\\             <wrap hi>"For each (scoped) business report, there are two tables: one that maps the report to source tables, and the other that maps the report to the Data Vault source for the information mart output."</wrap>\\  
 +When we take a closer look at that statement and divide it in its pieces we can quickly find out what is meant by this:\\  Business reports are usually scoped anyway in order to maintain its function, which basically consists of giving managers an overview to data in order to make rational decisions. And since the Data Vault Methodology enables us to produce any truth, we always have to be able to give details what a value, here called ‘requirement’, actually represents and how it has been computed as well as where it originally came from. To do so we can offer meta data for every value of every report: from which (data) table it came from. This is further explained in chapter 3.2.4, Technical Numbering. Now we have explained the first of these two tables, but what about the second data breakdown structure?\\ It simply gives meta information about every requirement used in the report as well, but this time it maps to information mart table, an already composed collection of //data// which can now be seen as information since an //information// mart adds meaning, so the usage of that term ‘**information** mart’ can be justified. The human itself by the way would now be the next step, we successfully reached the //knowledge// step. To be clear it would more or less be tacit //knowledge// because no human being will be able to break down his/ her decision making process into separate steps. This is basically due to the reason that //information// systems work by following algorithms while a human is more or less a black box, otherwise it would neither be tacit nor would there be a need for that deciding person. To finish that subject matter, the ultimate step, //wisdom//, is also executed by a human based on his composed knowledge, that we can call experience and the ability to objectify his decisions with giving reasons and fully understanding.\\ \\  
 +{{http://blog.ibsindia.org/wp-content/uploads/2015/06/Knowledge-Pyramid.jpg?500}}\\  
 +But now back to //DBS//: How does such an table look like? Well, like this: 
 + 
 +^          Requirements to Information Mart Tables Example           ^^^^^^ 
 +|Requirements to Target Map XREF\\ \\ Requirement Document:\\ \\ Table|\\ \\ MyDoc         | 
 +| Logical Name | Physical Name | Business Key | Passenger | Airplane Utilization | Connections | 
 +|Passenger Information| PASSENGER | | X | | | 
 +| Connections | CONNECTION | | | | X | 
 +| Airplanes | AIRPLANE | | | X | X | 
 + 
 +This offers full transparency and auditabilty which is often required in order to inform the deciding executives the best possible way. 
 + 
 +\\  
 + 
 +**Sources:**\\ (Definition, intellectual property and table taken from [[http://www.sciencedirect.com/science/book/9780128025109"|Building a Scalable Data Warehouse with Data Vault 2.0"]] written by Dan Linstedt and Michael Olschimke, page 69 - 72 since the rest of the World Wide Web didn't offer that massive amount contributed to that specific term as compared to ones used more widely)\\  
 +Picture of DIKW pyramid: [[http://blog.ibsindia.org/wp-content/uploads/2015/06/Knowledge-Pyramid.jpg|Link]]
  
  
concepts/data-breakdown-structure.txt · Last modified: 2020/08/20 13:27 (external edit)