Which of the following property of the core date warehouse layer of an enterprise data flow architecture uses common attributes to access a cross section of an information in the warehouse?
C. Drill across
Explanation
Drill across ?Use common attributes to access a cross section of information in the warehouse such as sum sales across all product lines by customer and group of customers according to length of association with the company.
For CISA exam you should know below information about business intelligence:
Business intelligence(BI) is a broad field of IT encompasses the collection and analysis of information to assist decision making and assess organizational performance.
To deliver effective BI, organizations need to design and implement a data architecture. The complete data architecture consists of two components
The enterprise data flow architecture (EDFA) A logical data architecture
Various layers/components of this data flow architecture are as follows:
Presentation/desktop access layer ?This is where end users directly deal with information. This layer includes familiar desktop tools such as spreadsheets, direct querying tools, reporting and analysis suits offered by vendors such as Congas
and business objects, and purpose built application such as balanced source cards and digital dashboards.
Data Source Layer - Enterprise information derives from number of sources:
Operational data ?Data captured and maintained by an organization's existing systems, and usually held in system-specific database or flat files. External Data ?Data provided to an organization by external sources. This could include data
such as customer demographic and market share information.
Nonoperational data ?Information needed by end user that is not currently maintained in a computer accessible format.
Core data warehouse -This is where all the data of interest to an organization is captured and organized to assist reporting and analysis. DWs are normally instituted as large relational databases. A property constituted DW should support
three basic form of an inquiry.
Drilling up and drilling down ?Using dimension of interest to the business, it should be possible to aggregate data as well as drill down. Attributes available at the more granular levels of the warehouse can also be used to refine the analysis.
Drill across ?Use common attributes to access a cross section of information in the warehouse such as sum sales across all product lines by customer and group of customers according to length of association with the company. Historical
Analysis ?The warehouse should support this by holding historical, time variant data. An example of historical analysis would be to report monthly store sales and then repeat the analysis using only customer who were preexisting at the start
of the year in order to separate the effective new customer from the ability to generate repeat business with existing customers.
Data Mart Layer- Data mart represents subset of information from the core DW selected and organized to meet the needs of a particular business unit or business line. Data mart can be relational databases or some form on-line analytical
processing (OLAP) data structure.
Data Staging and quality layer -This layer is responsible for data copying, transformation into DW format and quality control. It is particularly important that only reliable data into core DW. This layer needs to be able to deal with problems
periodically thrown by operational systems such as change to account number format and reuse of old accounts and customer numbers.
Data Access Layer -This layer operates to connect the data storage and quality layer with data stores in the data source layer and, in the process, avoiding the need to know to know exactly how these data stores are organized. Technology
now permits SQL access to data even if it is not stored in a relational database.
Data Preparation layer -This layer is concerned with the assembly and preparation of data for loading into data marts. The usual practice is to per-calculate the values that are loaded into OLAP data repositories to increase access speed. Data mining is concern with exploring large volume of data to determine patterns and trends of information. Data mining often identifies patterns that are counterintuitive due to number and complexity of data relationships. Data quality needs to be very high to not corrupt the result.
Metadata repository layer - Metadata are data about data. The information held in metadata layer needs to extend beyond data structure names and formats to provide detail on business purpose and context. The metadata layer should be
comprehensive in scope, covering data as they flow between the various layers, including documenting transformation and validation rules.
Warehouse Management Layer -The function of this layer is the scheduling of the tasks necessary to build and maintain the DW and populate data marts. This layer is also involved in administration of security.
Application messaging layer -This layer is concerned with transporting information between the various layers. In addition to business data, this layer encompasses generation, storage and targeted communication of control messages.
Internet/Intranet layer ?This layer is concerned with basic data communication. Included here are browser based user interface and TCP/IP networking.
Various analysis models used by data architects/ analysis follows:
Activity or swim-lane diagram ?De-construct business processes.
Entity relationship diagram -Depict data entities and how they relate. These data analysis methods obviously play an important part in developing an enterprise data model. However, it is also crucial that knowledgeable business operative is
involved in the process. This way proper understanding can be obtained of the business purpose and context of the data. This also mitigates the risk of replication of suboptimal data configuration from existing systems and database into DW.
The following were incorrect answers:
Drilling up and drilling down ?Using dimension of interest to the business, it should be possible to aggregate data as well as drill down. Attributes available at the more granular levels of the warehouse can also be used to refine the analysis.
Historical Analysis ?The warehouse should support this by holding historical, time variant data. An example of historical analysis would be to report monthly store sales and then repeat the analysis using only customer who were preexisting at
the start of the year in order to separate the effective new customer from the ability to generate repeat business with existing customers.
CISA review manual 2014 Page number 188