Let's understand why we should use the QVDs:

• Decreasing load on the source system: Most business intelligence (BI) tools fire a query on the data source when any selection or filtration of data is done. This causes heavy load on the data sources. When the data source is a transactional system, heavy load can crash the system. The QVD helps to avoid this, because it stores the same data in a compressed format. So, the Qlik Dashboard uses data from the QVD instead of direct data sources.
• Increasing the load speed: The QVD stores the data using a similar technique that is used by the Qlik Sense Engine to store data in memory. This enables Qlik Sense to load data faster from the QVD than loading it from the source system.
• Allows incremental load: The incremental load helps in decreasing the load on the source system and improving the load time. This is because in the incremental load, we fetch only the new records from the source system. The QVD helps in implementing the incremental load, because it holds the historical data and allows you to identify the last record of it, so that new data can be fetched from the source system.
• Allows single source of truth for data: Quite often, there are multiple copies of the same data available in multiple formats. The QVD helps to maintain the single source for all the analysis happening in various departments. For example, sales QVDs can be used by the sales team and the marketing team for their analysis.
• Helps in creating scalable solutions: When data grows, it becomes difficult to scale, but dashboards built with QVDs can be easily scaled-up. QVDs can be broken into multiple smaller QVDs that can be then used to scale up.

The syntax for creating the QVD is as follows:

Store [fieldlist from] TableName into QVDName;

Examples include the following:

Store Sales into Sales.qvd;

Store * from Sales into Sales.qvd;

Store SalesDate, Salesamount from Sales into [Lib:\ConnectionnameSales.qvd];

The following screenshot shows a script example:

The following screenshot shows the data load from the same QVD, along with the output: