A DATA WAREHOUSE is a collection of integrated, subject oriented, time variant and non-volatile support decision-making. (Bill Inmon) In
pills
* DW uses data from multiple databases,
* The DW ORIENTED TO SUBJECT . The data are organized by subject (or arguments) of the principal. (Eg Purchasing, Sales, Diagnosis). Each individual is represented abstractly as a ndimensionale cube (hypercube) . For any given subject there is a set of points (individual events) in an n-dimensional space. Each cell of the cube contains numerical measures that quantify the facts from different points of view. (Eg number of sales or total sales.) Each axis represents a spatial dimension analysis (which can consist of hierarchies of levels aggregation). A dimension can be any set of values, not numbers (eg Time, City, Product). Each point is then uniquely denoted by an n-tuple of dimensional values \u200b\u200b(coordinates). DW AND THE
* 'INTEGRATED UNIOCO. The data are the result of data extraction from the sources and organized by subject (or subjects) of the principal. Eg Sales, Purchases, etc. Diagnosis. Each subject then every single fact is uniquely denoted by an n-tuple of dimensional values \u200b\u200b(coordinates). * DATA
are not volatile and are dated (historical). The data warehouse can only grow with time. New data are entered in blocks are cut by appropriate sources of data. Eg data entry last month / quarter / year.
The set of data is therefore relatively static. There are no changes frequently (eg every hour)
Objectives:
- With a dw you can access all of the data, in a centralized DBMS
- Consistency and consolidation of data
- speed in accessing information
- Starting point for OLAP queries (set of techniques and software for interactive analysis of large amounts of data quickly. And 'the technological base of the DW. We can say that the main element of the architecture OLAP is the fact that DW contains data which when properly analyzed can provide important support to the decisions. OLTP systems (online transaction procession), primarily aimed at managing data online, provide data for the OLAP environment, we can say are a source of data. Between the two systems changes the type of user terminals, so users that perform read and write, for OLTP and OLAP for analysts).
OLAP tools are aimed at performance and achievement in the search query on a scale as large as possible. These are systems for data analysis. The OLTP systems are instead for data management. Can play a small amount Data on these and make a set of operations defined. Creating an OLAP database allows you to make a snapshot of information available in a certain time and then turn these items of information in multidimensional data. Performance of queries on structured data can get answers in a very short time compared with similar operations on other types of databases. One of the structures is the OLAP cube, multidimensional. The most famous is the cube that uses the schema "star" in the middle of this regular structure is formed a table listing the main elements on which the query will be built, and linked to this table the various tables of the "size" specifies how they will aggregate the data.
The life cycle of a DW from the definition of objectives and the necessary processing model, and then proceed with the design of abstractions, with subsequent refinement of the data needed, and finally comes to their integration into a single unit, the fact dw. The information contained in it are then categorized and divided into data marts that are released to end users. The term DataMart (literally store data) refers to a subset of the DW that contains the data extracted by sectoralisation dw and totalization. While
dw represents the entire information base executive of a company, the different data marts represent sections logic on it available to the various business areas. The data marts are undergoing a process of compaction in the light of sectoral objectives so as to be manageable and fast to use. A dw
can achieve important things: like the Drill-Down (allows you to add a dimension of analysis by disaggregating the data. It requires the formulation of a query), the Roll-Up (deletes a dimension of analysis, this can be done by adjusting the data of question), they Slice and Dice (select a subset of cube cells, obtained by slicing and cutting the cube itself.
A report is a call for specific formatted data. Reports can contain attributes and facts, filters to determine the amount of data used to generate reports and metrics to perform calculations on the facts.
pills
* DW uses data from multiple databases,
* The DW ORIENTED TO SUBJECT . The data are organized by subject (or arguments) of the principal. (Eg Purchasing, Sales, Diagnosis). Each individual is represented abstractly as a ndimensionale cube (hypercube) . For any given subject there is a set of points (individual events) in an n-dimensional space. Each cell of the cube contains numerical measures that quantify the facts from different points of view. (Eg number of sales or total sales.) Each axis represents a spatial dimension analysis (which can consist of hierarchies of levels aggregation). A dimension can be any set of values, not numbers (eg Time, City, Product). Each point is then uniquely denoted by an n-tuple of dimensional values \u200b\u200b(coordinates). DW AND THE
* 'INTEGRATED UNIOCO. The data are the result of data extraction from the sources and organized by subject (or subjects) of the principal. Eg Sales, Purchases, etc. Diagnosis. Each subject then every single fact is uniquely denoted by an n-tuple of dimensional values \u200b\u200b(coordinates). * DATA
are not volatile and are dated (historical). The data warehouse can only grow with time. New data are entered in blocks are cut by appropriate sources of data. Eg data entry last month / quarter / year.
The set of data is therefore relatively static. There are no changes frequently (eg every hour)
Objectives:
- With a dw you can access all of the data, in a centralized DBMS
- Consistency and consolidation of data
- speed in accessing information
- Starting point for OLAP queries (set of techniques and software for interactive analysis of large amounts of data quickly. And 'the technological base of the DW. We can say that the main element of the architecture OLAP is the fact that DW contains data which when properly analyzed can provide important support to the decisions. OLTP systems (online transaction procession), primarily aimed at managing data online, provide data for the OLAP environment, we can say are a source of data. Between the two systems changes the type of user terminals, so users that perform read and write, for OLTP and OLAP for analysts).
OLAP tools are aimed at performance and achievement in the search query on a scale as large as possible. These are systems for data analysis. The OLTP systems are instead for data management. Can play a small amount Data on these and make a set of operations defined. Creating an OLAP database allows you to make a snapshot of information available in a certain time and then turn these items of information in multidimensional data. Performance of queries on structured data can get answers in a very short time compared with similar operations on other types of databases. One of the structures is the OLAP cube, multidimensional. The most famous is the cube that uses the schema "star" in the middle of this regular structure is formed a table listing the main elements on which the query will be built, and linked to this table the various tables of the "size" specifies how they will aggregate the data.
The life cycle of a DW from the definition of objectives and the necessary processing model, and then proceed with the design of abstractions, with subsequent refinement of the data needed, and finally comes to their integration into a single unit, the fact dw. The information contained in it are then categorized and divided into data marts that are released to end users. The term DataMart (literally store data) refers to a subset of the DW that contains the data extracted by sectoralisation dw and totalization. While
dw represents the entire information base executive of a company, the different data marts represent sections logic on it available to the various business areas. The data marts are undergoing a process of compaction in the light of sectoral objectives so as to be manageable and fast to use. A dw
can achieve important things: like the Drill-Down (allows you to add a dimension of analysis by disaggregating the data. It requires the formulation of a query), the Roll-Up (deletes a dimension of analysis, this can be done by adjusting the data of question), they Slice and Dice (select a subset of cube cells, obtained by slicing and cutting the cube itself.