Data Warehousing is the process of aggregating structured data from one or more sources so that it can be compared and analysed for greater business intelligence. A data warehouse is constructed by integrating data from multiple sources which support analytical reporting, structured and ad hoc queries accompanied with decision making.
Flow in Data Warehousing:
Data that enters the data warehouse comes from operational environment in every case. Data warehousing provides tools for systematically organizing and understanding data for business executives so that they can make the strategic decisions. Many organizations have agreed that data warehouse systems are valuable tools in today’s fast moving and competitive world.
During the past several years, many organizations have spent a huge amount of money in building enterprise wide data warehouse. With the increasing competition in industry, the latest marketing weapon we have is data warehousing which keep the consumers engrossed by learning more and more about their requirements.
Data warehousing is a formal method for these techniques. The method of incorporating summary files in place of operational data is in essence what we call data warehousing. Some tools involved with data warehousing neglect the importance of modelling and building a data warehouse and focusses primarily upon the storage and retrieval of data.
Let’s go through a few advantages of data warehousing:
- We can perform new types of analysis with the help of data warehousing.
- We can integrate data from multiple sources.
- We can reduce the cost to access historical data
- We can improve the turnaround time for analysis and reporting.
- We can support ad hoc reports and inquire about them.
- We can share data and also allow others to access data easily.
- We can remove informational processing load from transactional oriented database system.
Some Key Features of Data Warehousing:
- Consistent view of data: Data that is required for business interface use is usually extracted from the operational system and then transformed to make it consistent to further load it into the data warehouse for analysis. This is the reason a data warehouse platform should be capable of extracting data from multiple source system and make it look alike a single platform for information.
- It enables your organization to model and create database designsfor data warehousing: A common requirement for a data warehouse is to prepare a database. It would be fair to say that online transaction processing systems completely rely on fully normalized schemas to guarantee data consistency and integrity. The online transaction system in most popular websites like Urban Ladder and eBay are also equipped with this feature of data warehousing.
- Data movement capabilities. The data warehouse is separate from the operational database systems that usually run daily business transactions. Thus, data needs to be regularly moved from one environment to the other. There are several methods and technologies for moving data which are as follows:
- Simple load and unload utilities.
- ETL functionalityfor extracting, transforming and loading data.
- In-memory functionality. If we use memory instead of disk for data storage and processing, performance can be improved. The various options include using an in-memory DBMSor adopting a data warehouse platform that delivers in-memory features.