Welcome to the In-Memory new school of thought with SAP HANA
I decided to postpone a new episode for the SAP HANA SQL for bizy geeks series, which I was concocting for a while, because I have a time critical suggestion to make, and if I wait too long to spill the beans, the window of opportunity will be over…
Here we go!
We are happy to sponsor the In-Memory Computing Conference, which will take place in Düsseldorf from June 4th to June 5th, and I would like to warmly invite you to join us. It would be a great opportunity to meet live, and further discuss development, business and SQL practices, and related quirks, in person, outside the blogosphere.
The deeper you go into the realms of In-Memory Computing, and of SAP HANA in particular, the clearer it gets that we are not simply looking at a new and very fast database technology with a fancy cache strategy. We are rather looking at a new school of thought around software applications…
As it happens with all new schools of thought throughout history, the new theories are not directly understood, they are often attacked, and in the best case, it takes a while until people learn how to take real advantage of them. Human nature tends to resist new ways, unless there is no other choice. I have seen more than one developer sighing with nostalgia, praising and worshipping the “good old” imperative development paradigm, ruled by the God of the ABAP heavens, and quietly cursing the fiendish, set-oriented, declarative HANA development paradigm, with its lack of good debugger tools, its need of nasty tracing strategies…
You could surely keep doing the same things you did before, develop SQL script based stored procedures à-la-ABAP, with their IF-ELSE-WHILE-LOOP statements, avoid the challenges, and with them, the need to change yourself. But then again, you would also miss the fun, the full meaning, and the full potential.
How can you bring HANA to optimize your SQL Script procedures in a compelling way, if you do not try to follow a set-oriented style, and build a highly parallelized Calculation Model as solution for the problems you want to solve?
How can you bring a compiler to optimize your SQL Script statements in a compelling way, if you are using too many IF-ELSE statements, or if you are not even trying to isolate those which are quite inevitable, perhaps in terms of a separate SQL procedure?
The right Calculation Model will not be built by SAP HANA automatically only by using the SAP HANA database, by pressing a magic button or by saying some magic words.
This is where YOU play a pioneering role: It is the work of great developers to look for great solutions.
Are you a great developer or a chicken *?
Co-Founder and COO
* Nothing against chickens, the poor beasts… They don’t have an easy live in their overcrowded farms…