Labels:text | screenshot | font | number OCR: Data Mapping Scenario Mapping Cardinality Explanation Cata Conditioning One held in the source systemi is split among many fields in the target System. In its most triv- al ingrancera concatenated held is parsed into its components For example Hotel Roomi Numa bes might be split into Roos Number"and" Sequence Number in the more complex situation, . Shgle hold has been used fur more than one purpose. For example WHEN.account number >1000THENVALD_X="Verdur Number ELSE ID_X= Customer Number, Whetyrnapping in the data warehouse these miles must be incorporated into the field's transformation antes. When reengineering the application, these types of rules should be eliminated, because the newr. application will hold the data at its atomic level. integration Many fields are combined into a single target fiekl Data elements that were distinct are now sven! as one and the semse This situation often occurs when data from multiple source applications is being integrated within the target database. a common occurrence when transforming data into the data warehouse. For example "Part No Part Not,"and"Part id' allt represent the same buisiness concept." Part Number" stasis One-to-one mapping between the source and target environments. Often, the old systems had it right and we only need to move the data over. We att like to see such cases, but we should ask ourselves flewrything were this easywould they need me?" Simplification The source data ciemerd doesn't map to any data element in the target environment This site ationceromonly occurs within the data warehouse when the source application is an cocamional system. Only data that is the basis of business managemera analysis wie map to data warehouse. data elements In a reengineering effort, each one of these cases must be evaluated carefullyein surmising number of such cases the datathat had been kepa is now conskiered useless Most aften th's data indudes things such as UPDT_AtG" that controlled processes that are nionger. meaninghil in the new system But you may also find fields such as Customer Hair Color" that were mit into the system at the request of someone who retired in, say. 1989 none The dara clement needed in the target environment is currently not available in the source ap- plication This shuation is an extremely common yet painhud one for data warehouse users Most source applications ere operational systems that heki only the data necessary to po toir those specific caerations. In many cases, the data required to manage the businesss health, isn't averable through the operational systems, The need to capture this data may lead to require- ments for enhancements to the source applications. When reengineering into a system, you can expect users to provide requirements for new data.In some cases, the new data will be acquired from external suites, such as Axcion or Dun & Bradstreet However, users can expect that some of the new deca will cintalit full values at the outset and for some time thereafter The impli- cations of mis missing data on the business environment, in terms of the data validation miles that the new application esiforces of the quality of the business analysis connected through the cata piorchomie, must be carefully cons deved. TABLE 1. Datamapping scenarios.
(1999).iso/dbprd/may97/morit101.gif){kind=small}
(1999).iso/dbprd/may97/de00f101.gif){kind=link}
(1999).iso/dbprd/may97/thomf101.gif){kind=link}