Data Warehouse: Data warehouse is a form of storage system (database) where large volume of data is stored in such a way that retrieving desirable information from the system is very easy and reliable. Data warehouse is stored in different location so that it doesn't collide with transactional database system which stores day-to-day information and answers the queries that are prerecorded in the database. Data warehouse system on the other hand provides solutions to sophisticated queries, which involves many computations to be performed at finer-level of granularity.

W.H.Inmon defined data warehouse as a subject (but not application) oriented, consolidated, time-dependent and non erasable collection of data which is mainly supported by decision management. From this definition, the following characteristic features of data warehouse can be defined as,

1. Subject-oriented (Not application-oriented)

2.Consolidated data

3. Time-dependent data

4. Non-erasable data.

Subject-Oriented: Data warehouse focuses mostly on important subjects as operational databases like producer, consumer, manufacturer i.e., data warehouse in not application-oriented. The major role of data warehouse is to support decision makers in making strategic decisions. It does this by performing data analysis and by applying data modelling tools. Data warehouse abstract unnecessary data which is not required in decision making and provide a precise view of data organized around a specific subject. Data warehouse doesn't maintain information about day-to-day transactions of organization but concentrate on the subjects which are critical to the organization.

Consolidated Data: Data warehouse is capable of retrieving appropriate data from heterogeneous databases (like relational databases, flat files) in order to make efficient strategic decisions. Because of the heterogeneity, data is stored in inconsistent manner. To confirm consistency and reliability in naming rules, encoding methods, different techniques such as data cleaning and data integration needs to be applied. Therefore, it is necessary to perform data transformation. data consolidation before transferring data from the operational system into data warehouse.

Time-dependent Data: Data warehouse database not only stores current information but also stores historic information about a particular transaction. In operational databases, though historic information is stored, it generates only preset information as these databases are capable of supporting only day-to-day transactions.

Data in data warehouse are archived as snapshots over historic and present time periods. In these databases, every data structures are time dependent i.e., they are directly or indirectly elements of time. This data warehouse approach is non trivial for design as well as for implementation stages of data warehouse.

The advantages due to time-dependent features in data warehouses are, 

• It enables analyzing historical data.

• It associates information of past data with present data.

• It provides better future prediction.

Non-erasable Data: This feature confirms that once data enters the data warehouse it remains static until a particular event is triggered. Data warehouse contains data which is extracted transformed integrated from operational database. The data is transferred from the operational system to data warehouse at regular intervals of time depending on the specification of the business. There are only two operations executed by data warehouse for accessing data.

(a) Data loading method

(b) Data access method.

Because of the above features, data warehouse can be considered as a consistent storage area that provides support for decision making and for analytical reporting.

Advantages of Data Warehouse

• Data warehouse is capable of storing and consolidating past information.

• It provides support for sophisticated multidimensional queries.

• It increases the performance of integrated database system as data from heterogeneous sources are extracted, preprocessed, cleaned, transformed into one unified data store.

• It doesn't use query driven approach (as it requires difficult filtering and integration techniques) instead uses update-driven approach where the data collected from various sources are consolidated and stored in warehouse for performing explicit data analysis.

• It has the ability of supporting management decision activities like managing relationships with potential customers, managing the cost of important assets, evaluating operation and searching for sources using which profit can be attained.

• It is capable of understanding the current business trends and making better forecasting decisions.