My encounter with the freight train called HANA

[holger_stumm2]

When first HANA occured in public, I was probably on the same side as everybody. I thought, that HANA is no new DB except a little bit faster than SSD based databases, something like a SAN on steroides. But what would that mean? Two, three times faster? What is the big deal? Maybe some backup stuff extra, that's it.

When I start digging deeper into HANA, with these wonderful classes at Hasso-Plattner-Institute in Potsdam and now the SAP HANA University with Thomas Jung, my viewpoint changed dramatically. More and more, the picture of a true revolution unfold before my techie eyes. First, yes, it is a game changer. No marketing bla bla, no power point ware.

Once you understood the concept and the stories behind it, you can really see the freight train called HANA running towards Redwood Shore. What a great clash that would be! (Disclaimer: I am a former SAP employee)

For me as an architect, a few important elements had to come together to form this drastic picture.

First, you need really big data problems. This is exactly the trend in the last years. We have seen SAP silo application, we mastered all kinds of MM, SD and Production and put the biggest companies on best practice.

Than there was Internet and Intercompany, everything was connected and the net was woven. The data from the proceses inbetween became focus.  Now, the corporation wants to tackle all the data and their informations between these networks and webs, the data from the cloud and the information shadows. The source data and the meta data, of goods, users and relations.

Then you need the technology to grab and store this. The cloud was providing these gigantic server farms, not as permanent investment, but also on a cost-coscious on-demand base. You need also the technology to read and write through all these data. Hadoop and Cassandra, the open source equivalents to HANA and Google, where showing the way, while suddenly Hasso was showing up in the game with HANA and the big picture of SAP in the middle of the Big Data hype.

Once you understood the concept of the column store and the relation to Googles Map/Reduce algorithms and understood the way, Hadoop for example and SAP are collecting data and provisiong the sets for HANA, you can tell about the huge impact this will have on the IT world and the SAP world.

Analytics are first to hit by the freight train. This is a low hanging fruit and the BI people are already in the middle of it. Big Data applications based on SAP structures and traditional table/SQL-Design are next.

The path from traditional SAP SQL data to HANA is powerful and will ease the way to transitions rather than migrations or release changes. Big data based on mashup of extracorporate data (discrete data and meta data) and SAP data are the next thing. The server farms of Hadoop, the commercial Google Map/Reduce implementations will meet corporate SAP computing. This will be the most exciting part.

Security, encryption, trust and certificates of data, traffic and communication between all these big technologies is another thing, If you look up, what Intel Xeon E7 processors have to offer and already implemented in the processor, you can tell why SAP has joined forces with Intel on these and many other hardware related topics. This by itself is an exicting journey into hardware and software.

Consulting will change. Cloud computing and the appliance approach are changing the vendor and the customer landscape to equal parts. Consultants are no longer needed on every site, on every problem. They will be needed on vendors sites as specialists, on customer sites as big data analysts, as landscape architects and as security specialists everywhere. Every site needs fewer people, but much more qualified (in depth knowledge) than today. Not to forget the programmers in Java, Javascript and SAPUI5. Server side script based power application, with mixed media UI.

This will be a colorful world.

And last year, I thought that SAP will become a boring topic in future.