One of the tools I’ve developed as part of the Lean Data Processing Framework is an optimized implementation of the QVD Segmentation technique. This technique is typically used in Qlik Deployments that deal with large datasets, and allows us to split a large table into multiple QVD files. For example, storing one QVD for each month of data, rather than storing the entire dataset in a single file. While this technique has always been around, and other implementations exist that perform the same operation, the modified version that I’m sharing today focuses on reducing the time it takes for the process to complete by avoiding over-processing, and reducing the motion of data required in the process.
The following 14-min video shows how the Delta Tags technique easily replaces another commonly used approach to incremental loads (based on variables) in QlikView and Qlik Sense, and demonstrates why the Delta tags approach makes the process more reliable while still being highly efficient and configurable. Another important aspect of this technique that is demonstrated in the video is how the Delta Tags approach works very well with segmented QVDs (a.k.a partitioned QVDs) in an incremental scenario, something that is not as straightforward to implement with other approaches.
Delta Tags – A new mechanism for efficiently keeping track of incremental reloads in QlikView and Qlik Sense
When working with large datasets in QlikView and Qlik Sense, it’s important to ensure the backend data processing jobs (Extracts and Transforms) perform efficiently, by following best practices to reduce execution time and keep resource usage at a minimum. One of the best practices recommended for efficient data processing is implementing an incremental load strategy, [...]