Over the past weeks I have been researching SAP HANA. As a software engineer it is an intriguing technology with some interesting potential. Sometimes my software engineering background is an asset when learning something new. However in this case it might actually be a bit of a barrier initially. I find myself wanting to cram the HANA concepts into my classic relational DB mindset and concepts of how “things are done”. That works to a point, but it became obvious to me early on that continuing that way would lead me to miss the true HANA paradigms and not leverage it to its fullest.
Today while sitting and Barns & Noble bouncing between different SAP HANA articles on my iPad, I had an epiphany which I think will be a breakthrough for me. While reading and working with HANA I keep wanting to make a classic snowflake schema with my HANA tables, setting fact tables in the “middle” and placing dimension tables “around” it. Then it hit me, in HANA you don’t need to do that with the raw tables. You can leave them in their “natural” state, then use the different HANA views to make the snowflake or star schema by placing attribute views (dimensions) around analytic/calculation views (facts).
This probably seems obvious to some, or like a basic concept to those who have worked with HANA already in the short time it has been out. And it true, now that I see it, it’s quite obvious. Yet this is a mental shift in the way I think of these elements which will help me in my research and road to certification. Maybe my sharing this will help a few other who had the same mental block as me.
I have comments turned off right now, but please feel free to share your own SAP HANA epiphanies with me on twitter (@rwbDev), or simply heckle me for not understanding this concept sooner.