Is There Anything Left To Be Said About SAP HANA – Yeah, There Is

I have written a lot about HANA, and have participated in several podcasts, webinars and so on about HANA. And I am probably the least prolific person around on the topic. SAP and its influencers have all done a lot to spread the word to the extent that I was joking on twitter recently that I wonder if there is anything left to be said about HANA.

I guess there is – about what is going to happen next about HANA.

Apart from the stand alone HANA option and BW on HANA that are both in GA, a lot of things are in the pipeline from what we have heard from SAP – ECC on HANA, CRM on HANA, and every other business application from SAP on HANA, the HANA App cloud and so on. There is also a lot of activity going on about HANA start ups. We know this – but what remains as a question is what will SAP prioritize when it comes to HANA ?

We have heard Vishal Sikka state that ECC running on HANA in some shape will come out in Q4 this year. I know it is a topic close to Vishal’s heart – and for good reason. Having heard Hasso on the topic too multiple times, I know this is a big priority for him as well. I think the primary reason to do this is to prove to the world that HANA can take transactional workloads as well or better than Oracle and other traditional DB vendors. It is a coming of age of sorts for HANA. So this definitely has merit as a worthy goal for SAP to aim at. But is this a good commercial goal for SAP?

HANA apparently has about 600 customers for what is available today as functionality. Not all of them have projects on HANA, but probably will have one at some point. Now, it has only been out for a limited time – so we have to give it time. For a 16000 strong install base, HANA has barely skimmed the surface. Probably it will sell a lot more – and it is a safe bet, since BW is not always a mission critical system, and more customers can be persuaded to buy HANA for BI purposes even though it has only gone into GA recently. There is enough value to offset cost for most BW customers.

However, that is not the story for ECC. Most companies will come to a screeching halt if anything at all goes wrong with ECC implementations. Although to a lesser extent, so is the case with CRM. So I am curious as to how many customers will make the leap into ECC or CRM on HANA in near future. I am sure there are a few that always will buy into SAP’s vision and jump into it. But the vast majority of SAP customers might take a conservative stance on changing their database for a production ECC or CRM instance.

As I and many others have pointed out before – just by porting ECC from some database to HANA, things won’t get incredibly fast by default. The reason is that most of the bottleneck is in the ABAP layer. Unless all that code is rewritten, the schemas simplified and so on – it is difficult to imagine ECC consistently getting a huge boost from Hana. I am sure SAP will probably rewrite some code to make use of HANA, at least for long running batch jobs. But even in that case – the best case I can see is most customers using it as a secondary database for accelerating some parts of ECC at best. Add to this the efforts of all the other DB vendors to attack HANA to protect their turf. So all things considered, running ECC etc on HANA is at best a technical triumph, but not a sure shot commercial success in itself.

Then there is the HANA App cloud. When I first heard about it from Vishal, I was super excited at the potential. But not a lot of action has happened for customers to jump into it. But I am a bit confused on the whole cloud strategy at SAP now. There is the big cloud division under Lars Dalgaard that has BByD, LOB On Demand solutions, SFSF etc. Then they bought Ariba – and I am curious to see where that will fit in. Will Ariba folder under the existing cloud organization? I doubt it – but curious to see is it does. And then there is the Hana App cloud, and netweaver cloud. Where does all that fit? will it stand alone ? Will it fold under Lars’ organization? Why does this matter ? Only because now SAP has so many disparate technologies and data centers, that I am not sure they have the time and effort at their disposal to get synergy out of all of this – in data centers, in applications, and in platforms. Sure we can assume that user experience will converge at some point, but if the back end is heterogeneous, it is hard to drive down costs. I hope I can get some clarity from the SAP people next time I meet them.

Many years ago – SAP made an all out effort , though not quite as impactful as current efforts on HANA , on Netweaver. Netweaver was supposed to be the super platform that would make everything work together seamlessly, and with great performance and low development effort and all that good stuff. As it turns out, it was only a moderate success. Not many people – except some die hard SAP shops – trust netweaver as the platform to run non-SAP stuff. With HANA – SAP has a chance to correct that, and make a big platform play.

Hana as a Platform is where I think SAP has biggest bang for their buck. I also think that is where SAP customers and the developer ecosystem also has the biggest bang for their buck. I am not sure if SAP sees it that way – but surely there are people there who are exploring that. Look at all the start ups, the interest in HANA during techcrunch and so on. These are not the traditional SAP ecosystem people. Their needs are different, and they have no loyalty from past that will keep them tied to SAP . And SAP has a relatively short window to keep their interest and make them loyal.

But there in lies the problem – there is only so much money and time SAP can throw at HANA, despite their deep pockets and excellent commitment at all levels of the company. The needs of HANA as a Platform are not always consistent with the features it needs to be an excellent database for ECC and other business suite products. If SAP tries to do this sequentially by getting HANA ready for business suite first, and then think of Platform plays, they might not have enough of interest from the ecosystem by the time they are ready for pltforms. Of course when I say platform, I don’t mean just the technical features – I mean everything from Go to market, pricing, support – the whole enchilada.

There is a blogger meeting with SAP coming up in few days – I will try to find answers to these while I attend those sessions, and post a part 2 if needed.