You are here

DAMA Sydney March 2015 Meeting

Date: Tuesday, March 10, 2015

Location: Sydney (Australia)

Denodo will be present in DAMA Sydney offering information about how data virtualization can help Australian companies with local's data retention law's compliance.


Topic 2 - Data Virtualisation

Specialised data modeling techniques for data warehousing made a profound impact on agility and extensibility of management information solution delivery. One of the next big things that blossomed on this was a wave of improvements in ETL delivery using DWH/ETL automation. Now we are at the point where we can take the next step: data virtualisation.

If we can (re)generate all ETL and reload the entire enterprise data warehouse at any time and with predictable/deterministic outcome, what does this mean for the relevance of ETL software? In the context of contemporary data warehouse architecture and modelling techniques, a key technique that is used to improve delivery of usable and actionable information deliveries in the form of data marts is data virtualisation: create a view on your core data warehouse that simulates the facts and dimensions. This presentation takes this concept one step further: if we have all the information and processing power available, how can we improve time to value even more? The answer is: by applying data virtualisation to the core data warehouse model.

Approaches such as anchor or data vault modelling are all about improving time to value, while still making sure a solid, predictable and extensible data foundation is deployed under the covers. These approaches aim to combine short delivery cycles with proper information management concepts, and fast deployment of usable and actionable information deliveries are a key point in this approach.

However, creating these foundations themselves can still be quite challenging. Even though much of the development can be automated there is still room for error, changes in interpretation and rework. This can be omitted to a large extent by using virtualisation techniques to quickly model and demonstrate key elements of the solution before committing to development.

This presentation explains how this works, and how this can be used to improve delivery and in many cases improve data query performance as well. A part of the presentation is reserved for a live demonstration of applying ETL techniques to support visualisation.