Rouse (n.d.) described facts warehousing as a federated repository for all the records that an agency’s diverse business structures gather, it emphasizes the size of statistics from numerous resources for useful analysis and get admission to, but does no longer generally start from the factor-of-view of the end user who may additionally need get admission to specialized, every now and then local databases but the latter concept is referred to as the facts mart. In a discussion of Oracle about Oracle9i facts Warehousing guide, a facts warehouse is a relational database that is designed for query and analysis in place of for transaction processing. It carries ancient statistics derived from transaction records, however, it could include statistics from different sources. It separates evaluation workload from transaction workload and permits a company to consolidate information from several sources. Sampagar (2016) explained the steps which are concerned with a top-down approach. step one is to extract the information from the numerous supply structures then, extracts could be loaded and verified inside the level location. The validation is needed to make sure the extracted facts are correct and correct. ETL equipment may be used to extract and push to the records warehouse. the second step is to extract the statistics from the facts warehouse on a regular basis inside the degree region. At this step, numerous aggregations, summarization strategies have to follow to extracted statistics and loaded lower back to the records warehouse. finally, once the aggregation and summarization are finished, statistics extract from numerous records marts and apply a few extra transformation to make the records structure as described. under diagram depicts how the lowest-up approach works. which is likewise called dimensional modeling or the Kimball methodology. Sampagar (2016) additionally defined the steps which can be involved in a bottom-up method. the first step is the extraction of statistics from the distinctive supply machine into the level location where it’s miles processed and loaded into the records marts because the facts flow in the backside-up technique starts on the supply systems. After records marts are refreshed the modern information is extracted another time within the degree area, differences are implemented to create statistics into the records mart shape. The facts are then extracted from information Mart to the staging region then the statistics can be aggregated, summarize and lastly, loaded into EDW after which made available for evaluation of the quiet customers. The below diagram depicts how the pinnacle-down approach works.