Data warehousing (or data storage) is a fundamental activity in any corporate environment. Data is one of the most important assets that any organization has, and we have already commented on numerous occasions to what extent it is crucial to have an adequate data management strategy . Conveniently managing the data and information housed in data warehouses requires, first of all, having a certain volume and variety of data that justify the adoption of an information governance system . Obviously, without raw materials to manage, any strategy aimed at this goal is nothing more than smoke. A condition that, despite being absolutely necessary, is in no way sufficient. Data warehousing : how to facilitate data analysis and management One of the main purposes pursued by a data warehouse , beyond simple data storage, is to serve corporate demands for analysis and information .
Therefore, when conceiving and designing a database, it must be taken into account that it is not only a space (physical or virtual) in which all the data of an organization is housed indiscriminately and in disorder, but of a space that must comply with very specific and defined characteristics: Orientation : the data always refer to certain aspects or themes of reality. A reality that can be past (historical data), present (operational, functional data...) or even future (data related to possible scenarios), but which in any case must serve to orient (order) the stored data according to its referent. or referral. Variability – Data warehouses are not static at all; On the contrary, they are constantly being fed with new data and information. A database that India Part Time Job Seekers Phone Number List does not allow correct reception, fluidity, updating and renewal of the same is obsolete from the moment of its creation. Involatility : a data warehouse is variable and flexible, but at the same time it must avoid volatility, that is, ensure the maintenance and integrity of the query data to guarantee that these same data and information are not altered with the queries that are made. , and that will be available for future requests for information. Integration : Closely related to data orientation, this characteristic of data warehouses guarantees the integration of new data from different sources in a consistent format, avoiding inconsistencies, errors and conflicts of all kinds. Having enumerated these basic characteristics of data warehouses , which we have defined as storage spaces oriented to the analysis and management of data and information, it remains to define the most convenient data warehousing methodology for each type of organization , that is, the most convenient way of store this data and information for later use.
Traditionally, there are two types of information governance methodology or strategies , choosing one of which will decisively determine the design of the corporate data warehouse : Descending or top-down : it is the methodology defined by one of the first researchers in the field of data storage, Bill Inmon, characterized by an information management strategy that is based on corporate needs to structure data and information in function of these. Ascending or bottom-up : methodology defended by Ralph Kimball (another leading name on the issue), consisting of considering the data warehouse as the sum of all the data marts of an organization. Its structuring, therefore, does not depend in the first instance on the specific information needs of the organizations, but is based on the same data and its characteristics or typologies. IBM one of the business intelligencesolutions offered by IBM , helps companies deliver quality data for their big data , business intelligence, data warehousing , application migration and master data management projects. To expand this information, we recommend reading the free guide 10 keys to defining your corporate data management strategy , which explains, among other matters of interest, the keys to defining the most appropriate data warehousing and data management strategy for each type of organization.