Technical Requirements

This section will look into the technical requirements that need to be considered when building a report. The purpose is to inform you about the different data connections that can be made and their performance.

Performance: Data Size and Structure

Tableau is capable of processing large volumes of data, but performance sits on a curve: the larger the dataset, the more computational power is required to access and process it. There is no fixed rule for when performance will meaningfully decrease, as this depends on a complex combination of factors, including the specification of the machine running the query (one with lots of resources, such as RAM, can handle greater quantities of data). It is fair to say that a data source with dozens of columns will be processed slower than one with a handful of them; similarly, a source with millions or even billions of records will be less performant than one with a few hundred.

There are stricter limitations for data sources hosted on Tableau Server or Tableau Cloud rather than a local machine; for example, joins and relationships cannot be established, only blends. These are covered in more detail in Chapter 8, Publishing and Managing Content.

It is worth noting that Tableau generally prefers data that is long rather than wide in structure: that is, Tableau can handle more records better than it can handle more fields.

Data Format and Compatibility with Tableau

Users should be sure that a connector exists natively for the given data source type. This can be a type of file that exists locally on the computer such as an Excel file.

Users should consider whether data is accessed live or saved as an extract – that is, whether the data is a saved snapshot, such as an extract, or whether it would run on a real-time basis, such as a live data source.

The description and limitations of these connections will be explained further in this chapter.