We love data, and what can be done with it. But even we shake our heads at the exuberance of terms used to talk about HOW data is stored and accessed.
The bad news is it’s not just jargon that you can ignore; the approach you take really will make a difference to how well you can use the data you have.
But here’s the good news. It’s not nearly as hard as you might think, to get your head around these approaches.
A data lake is the whole universe of your data. The term stems from the idea that you go fishing in it and land some valuable …er…insights. By definition, a data lake is unstructured and contains data in its rawest form. The responsibility for layering meaning and comparability onto it is left to the person or tool doing the querying. The technical term is schema-on-read.
A data warehouse is by definition structured; it contains data where the relationships and meanings have been considered and designed in advance, and are enforced at the point of storage. It puts the responsibility for making sense of the data onto the data warehouse designer rather than the person querying the data. The technical term is schema-on-write.
A data mart is easy. You can forget about it! It’s just an old-ish term for a subset of a data warehouse, for example purchasing / sales / human resources data. The underlying technologies have become so powerful now that the need for data marts no longer exists.
And what about Datitude? Datitude's Data Lakehouse is a hybrid solution that is both data lake AND data warehouse. We process raw data into standardised, highly validated and user-friendly data models, making standardised reports, insights and extracts fast, reliable and repeatable...
…we keep all the raw data too, so that it can be queried directly by the data scientists among you to gain new insights…which can then be standardised for end-users into the data warehouse.
You could picture your data warehouse as a living-breathing-highly-organised island in the centre of your beautiful data lake. And of course, your lake is in the clouds.