Looking back at Developer Deep Dive on SAP HANA, Part II

Recently I heard that the SAP TechEd Live interview “Developer Deep Dive on SAP HANA, Part II” with Thomas Jung, Graham Robinson and me became “Interview of the Week” at the SAP Teched Event place on SCN. If you didn’t watch this video I recommend you to do so:

https://youtube.com/watch?v=Uojl9Fjbi6Y

When I look  at the interview the information we provided have been quite accurate:

• Tom Jung explained about HANA Extended Application Services (formerly known as XS Engine), the new features of NW 7.4 and why ABAP developers should start to extend their knowledge of Open SQL to the whole SQL language.
• I talked about our Business Suite on HANA activities, my personal view of the evolution of the ABAP language and explained the HANA is more than a database – it is a calculating engine which is an enabler for predictive analytics.

These topics are still correct and when I look back I have to add the following information:

• It seems to me that many ABAP developers still try to solve too many tasks in the ABAP memory. Of course accessing ABAP memory is faster than accessing HANA memory but databases are optimized for calculations (summation, aggregation and so on) and the same holds for HANA. I checked many programs (custom and SAP development) within the last months and for me it was surprising that using more Open SQL features can improve the runtime of many applications on “anyDB” and perform on HANA even better.
• In general prediction of runtime on HANA is very difficult. Different transactions behave differently with different customizing, different application data and different HANA sizes.
• HANA is really more than a database – besides R, Business Function Library and Predictive Analysis Library it supports IMSL and I think much more will come. When I want to inform yourself about what the SAP HANA people are working on then look at their publications starting with Franz Färber at DBLP Computer Science Library with Franz Färber in scientific publications on journals and conferences. Then you should look at the work of people who published joint papers with him and so on. If you want to know what the future might bring you should have at look at the SAP HANA Research Group at Walldorf: http://scn.sap.com/docs/DOC-27051 Some people are working on really interesting topics like Graph Databases. Many people are working on this (Graph Based Data Mining or Facebook Graph Search for example) and I am curious about the result of SAP’s activities.

But back to the interview above: We tried to explain that HANA is more than a database cache and an accelerator for existing FI-CO and other transactions. It has really strong and unique features which we can use to build next generation applications.

And again: I don’t expect that ABAPer will have to become SQL and HANA experts overnight but they should learn about strenghts and unique capabilities of SAP HANA and learn how to use them,