Power BI is a self-service BI platform, launched by Microsoft ahead in the race, during the 2010 Microsoft Excel product release. It was initiated with the name Power Pivot for Microsoft Excel. However, the initial launch did not pick up with the BI end customers, so, Microsoft gathered user feedback, and carefully constructed the current version of Power BI.
The following diagram depicts the Power BI architecture:
As described in the preceding diagram, Power BI has two key layers, namely, Excel, and Office 365. Pretty much any type of input data source is supported by Excel. At the same time, Power BI report has common platform support like Windows, web, and mobile-based end user experience.
Power BI is an evolution of the add-ins previously available in Microsoft Excel, like Power Pivot, Power Query, and Power View. Let us explore the functionality of these components.
PowerPivot is placed as a data analysis tool. For customer convenience, it can be easily used within the Excel product rather than a standalone tool. Moreover, this tool helps the end customer to ingest the data literally from any kind of data source without much restriction. In turn, it adds a degree of flexibility in the ecosystem. In terms of output, PowerPivot generates end custom reports based on data insights, data analysis, collaboration, and so on.
Power Query is used for the purpose of data discovery; it can be easily embedded in an Excel add-in. It helps the end user to reshape the data and to combine the data coming from different sources. Power Query is supplied as a plug-in product from Microsoft Power BI self-service solution. It is considered the common ETL (Extract Transform Load) tool component. It has an intuitive and interactive user interface, which can be used to search, discover, acquire, combine, refine, transform, and enrich the data.
Power View is a vital output component. It enables the end user to build interactive data exploration and visualization. The Power View product promotes intuitive ad hoc reporting using enterprise data sets. As data visualizations are dynamic in nature, it is easy for the end user to produce a single consolidated report.