Data Integration approaches -ESB, EAI, EII and ETL
ETL: Extract Transform Load is the process by which data is extracted from sources systems (files, dbms etc) into a data warehouse. Data is migrated by extract, transform and load process as per the predefined data extraction logic. ETL process is batch, righttime or on-demand driven.
EAI: Enterprise Application Integration is the mechanism by which enterprise applications are integrated primarily by message managementData (information) integration. EAI ensures that information in multiple systems is kept consistent. It also makes sure that processes across applications are also integrated. Data is migrated by application messages which work based on the predefined application logic and navigation directions. EAI processes are usually event triggered.
EII: Enterprise Information Integration is the ‘on the fly’ data integration method where data is extracted from different sources and combined on the fly. EII can be used as a cost effective alternate solution for BI/ETL/DW. But EII cannot replace a BI/DW solution based on ETL. It is good for smaller amount of information integration and can be used for operational reports whose source will be change periodically. EII server engines make use of the metadata to understand and integrate data from different sources. Processes are triggered by online request. Data is extracted by queries generated by the engine based on the metadata.
ESB: Enterprise Service Bus