ISDS 2001 Lecture Notes - Lecture 5: Data Integration, Bill Inmon, Data Migration
21 Oct 2014
School
Department
Course
Professor
Document Summary
Application 2. 1: customer retention, reduce churn, cost reduction, customer acquisition, social networking => become opportunities, highly targeted data analytics, become better than competition, do not need to know. Data source => etl process => meta data enterprise dw => access data mart => application (replication) Improve patient care, provide data driven decisions, provide data on patient needs, hospital practices, patient outcomes, improve efficiency, profits, reduce mortality: define standard care for septic patience identify who might likely become worse, efficiently implement standard of care. 3 tier: client workstation, application server, database server. 2 tier: client workstation, app & database server. Issues to consider when deciding which architecture to use: which dbms, parallel vs partition, which data migration tool, what tools to support data. Integration 3 processes: data access, data federation for mergers, change capture. Issues affecting the purchase of an etl tool: data transformation tools are expensive, long learning curve.
