It is inevitable that each transaction, the organizations involved are fully aware of all relevant transaction data so that a decision can be arrived at without any ambiguity, with full transparency and signed with mutual satisfaction. In today’s world of cloud computing, the amount of data stored in different formats is mindboggling. Natural result is that only the data needed, with the riders around, you should press. The warning indicates that the organization is able to store data in a specific format and the company can be accessed within a few seconds the obvious additions of database or data warehouse.
In computer jargon , Extract, Transform and Load (ETL) refers to a process in database usage, especially in data warehousing that involves: Extracting data from outside sources.
Transforming to fit the operational needs of the highest quality level. Loading the final target (operational data store, mart or warehouse) The Extraction Extracting data from various internal and external source systems, structured and / or unstructured is the first stage of the ETL process. It can be very tricky, since only the relevant data at one point in time and correctly extracted organize the results. A simple request is sent to the system of origin, with a connection at home, message queues, open database connectivity (ODBC) or Object Linking and Embedding, Database (OLE-DB) middleware. Most data storage that collects data from various sources. Every bit of data possible source format, which can in relational and non-relational database structure. The purpose of this section extraction is to convert all the data in a specific format for transformation processing. Some ETL tools that can do this automatically. The data is then moved to what is called staging area.
The Transformation Once staging data are available in this area, all on one platform and one database. It becomes easy to combine tables, filtering and classification of data using custom attributes. A set of rules or functions to be applied to data obtained from sources to obtain the data for loading into the end target. Usually, several transformations may be required to meet the business needs and technical target database, such as translating the values of sort codes, apply simple or complex data validation, etc. The load Loading Data end target, usually a warehouse, as well as the fact / dimension tables. From there the data is collected and put into datamarts or cube as felt appropriate. Due to the requirements of specific organizations tend, this process may vary widely.
The ETL process is also referred to as the process of integration of data. Management of ETL processes such as data transfer, data management, data cleansing, data synchronization and data consolidation.