A short journey with a pricey tech
Use Cases and Deployment Scope
I used TIBCO Data Virtualization to build a federated query engine extracting data from several different sources and serving it up to a semantic layer to be used by the business.
Pros
- It has ability to connect to a variety of platforms
- It offers a central management console for admins accessible on web browser
- It abstracts the complexity of ETLs from the end users of data who consume it to drive decision making.
Cons
- TIBCO Data Virtualization does not offer out of the box it's own caching layer. It has to be augmented with some other database technology to cache data for efficient querying.
- TIBCO Data Virtualization is very tedious to maintain when it comes to managing granular permissions on objects and artifacts.
- TIBCO Data Virtualization by default tends to process the data in its own server memory instead of delegating it to the backend systems where data comes from. This sometimes leads to a server crash if not handled properly.
Likelihood to Recommend
TIBCO Data Virtualization is well suited for customers who are challenged to deal with extracting data from dozens of different sources and systems, and do not have the time and liberty to hire data engineers and/or ETL developers to write dozens or hundreds of complex ETLs.
However, there are situations where TIBCO Data Virtualization severely underperforms, and those are where we are dealing with large volumes of data, in tera bytes or peta byte scale system. For example, a messaging queue which sends 200 million messages every hour will choke TIBCO Data Virtualization if the technology is chosen to route the data.
