 welcome to a session on dimension modeling. This is Dr. Anita Poojar, Professor of Computer Science and Engineering Department from Volchin Institute of Technology, SolarPore. These are some of the prerequisites. Learners are required to have the knowledge of ER modeling. They should be familiar with the concept of OLTP, that is online transaction processing and OLAP, that is online analytical processing. At the end of this session, learners are able to compare ER modeling and dimension modeling, design fact table and dimension tables for given business processes. Here are some of the definitions of dimension modeling. According to Kimbal, dimension modeling is a logical design technique used for data warehouses. Now data warehouse is a huge central repository that collects data from multiple sources such as OLTP systems, ERP systems, flat files and so on. A dimension model is a database structure that is optimized for online queries and data warehousing tools. Dimension models are designed in favor of end users for reading, summarizing and analyzing numerical information. Now let's see what is dimension modeling. Every business process consists of two types of data, numeric data and descriptive data. Numeric data are also called as facts. They are the quantitative information about the business processes. These facts, they help to measure the performance of business processes. Descriptive data are also called as dimensions. Dimensions provide the descriptions or they provide the textual attributes describing the facts or the business numbers. Now let's take an example of retail shop. It consists of following data, date, product, quantity, store, transaction number, unit price of the product and sales amount. Now all this data can be categorized into dimensions and facts as follows. Date, store, product and transaction numbers. They are the dimensions that describe the business numbers or facts such as quantity, unit price and sales amount. These facts help to measure the performance of various business processes. Now dimensions and fact tables are needed to be stored in the database table structures and there should be some methodology for storing the data into the database tables in such a way that it provides easier and faster retrieval of data for various analysis purpose. Now let's see why dimension modeling is so popular. Dimensional modeling has three advantages. The first advantage is simplicity. That is, the data that is stored in dimension model is very easy to be understood by end users and they are readable and they help the end users to easily summarize and analyze the data. Separate models are used for each business process. The second advantage of dimension modeling is performance. Dimensional modeling provides enhanced query performance because the data that is stored in dimensional modeling is optimized for online query processing and various analysis. Dimensional modeling is also scalable. That is, it is flexible. The number of dimension tables and fact tables can be increased as per the increase in number of business processes of an organization. Dimensional modeling is also cost effective because the data in the dimension modeling is a denormalized data. That means number of tables required to store the data in dimension modeling is less as compared to as compared to the ER modeling that stores the data in number of tables especially in the third normal form. This dimensional modeling is widely accepted for analytic data. Now pause the video for a while, think and write which data modeling is used in OLTP and OLAP systems that is online transaction processing and online analytical processing. So it is like this. ER modeling are more suitable to be used in OLTP systems because OLTP systems capture data from day to day transactions and do the data administration that is they store the data, retry the data and modify the data. Dimensional modeling are more suitable in OLAP applications because these applications are used only for online query processing and analysis of data. That is dimensional modeling are more favorable for end users whereas ER modeling they are not suitable for end users but they are suitable only for data processing and administration. Now let's compare dimensional modeling with ER modeling. Dimensional modeling stores the data in facts and dimension tables. ER modeling is a graphical modeling that shows the data in terms of entity and their relationships. Dimension modeling is a logical design technique that is used for data warehouses. ER modeling is also a logical design technique but to remove redundancy and to show relationships between data. Dimensional modeling is designed to support end user queries but ER modeling is more useful for capturing the transactions data and database administration. It is not favorable for end user queries or end user delivery. Dimensional modeling provides physical model. Relational modeling provides both logical and physical model. Dimensional modeling uses historical data whereas ER model uses current data. Dimensional modeling stores data in denormalized fashion whereas ER modeling stores data in normalized fashion. Dimensional modeling can be mapped for creating schemas. ER modeling cannot be mapped for creating schemas. The size of data in dimensional modeling ranges from gigabyte to terabyte which is larger than the size of data in ER model which is from megabyte to gigabyte. Dimensional modeling provides view for business users whereas ER modeling provides view for data processing and data administration. The data storage in dimensional modeling is non-volatile whereas the data storage in ER modeling is volatile. To summarize this session dimension modeling is a set of guidelines to design a database stable structure for easier and faster data retrieval to support end user queries. So, dimensional modeling is a logical design technique for data warehouses. The advantages of dimensional modeling are its simplicity, faster query performance and cost effectiveness. So, dimensional modeling is made for end users. These are some of the references. Thank you.