Skip to Content

I had the chance to attend SAP’s database and mobility analyst event on April 10th, 2012.  Three things caught my attention as I listened in the audience:

1- SAP seemed to finally get that folks who have been considering HANA need some benchmark information.  Just saying “it’s fast” or “#x speed increase” isn’t quite enough.  Vishal Sikka got up on stage and delivered a slide with some details:

  • On a 100TB benchmark, with 100B records (sales and distribution, looked like BW)
    • Ad-Hoc analytics- 800ms-2s
    • BW analytics- 300-500ms
    • 20x compression
  • I snapped a picture of the slide, but it’s WAY too blurry.  Fellow SAP Mentor Vijay Vijasankar was in the audience and his is much more readable. And his blog about the event also has some great details that I won’t try to duplicate.
  • Announcing pricing for HANA was way overdue and nice to see.  Any SME who wanted to know what it would cost to run HANA now have some idea, as well as BW and ‘HANA Enterprise’ now have a public pricing scheme.  No actual numbers for HANA BW or HANA Enterprise, only that it’ll be charged in blocks of 64GB.  Also discussed was that there’s no license cost for dev and QA environments, just hardware. If you’re running a separate Dev/QA/Prod environment setup, you’ll only pay the license cost once and can install on all three servers. 

I’m not a HANA expert, but as a more technical user- the speed and cost numbers look interesting. Obviously, ‘your mileage may vary’ should apply. But it was nice to see some actual data points from SAP regarding HANA.  I am guessing that we’ll see a repeat of these numbers at ASUG Annual Conference/Sapphire next month.  If more detailed numbers aren’t put out in May, I’d hope for TechEd that SAP will come to bat with a larger cache of benchmark details.

2- The purchase of Syclo was news to me, but then again I don’t have any ‘skin’ in the mobility game- just a gadget infatuation.  But what interested me was Sanjay Poonen’s answer to the ‘why did you purchase them’ question during Q&A: SAP bought Syclo because of the apps.  It looks like the mobile marketplace continues to be defined by ‘apps’ and not having a core group of them was something SAP saw as a deficiency.  So now, SAP has two mobile application development/delivery platforms between Sybase and Syclo. I am sure that folks watching the enterprise mobility space are going to be dissecting where we go from here. (My money is on SAP Mentors John Moy and Kevin Benedict)  On a more ‘local’ note- Syclo is a regular sponsor of the Philadelphia ASUG Chapter and provides some pretty good mobility sessions. I hope to see that continue, even as SAP and Syclo formalize the purchase and deal with the inevitable transition.

3- “To be the #2 database by 2015”.  Those are some lofty goals that SAP has put out into the market recently and Oracle and Microsoft haven’t been sitting on the sidelines waiting to be lapped.  I would think that even with the cordial relationship they have, IBM isn’t going to let DB2 get passed on the hard shoulder by SAP.  Competition is a good thing and a little confidence and rhetoric usually makes for some decent hay.  But what struck me was Vishal’s answer in the Q&A that while the #2 goal is nice, they’re not going to sacrifice business value to get there.  I think that his comment that SAP will look to be the fastest growing database vendor makes more sense. As a customer who’s experiencing some feature loss between SAP BusinessObjects products and versions, I would rather SAP not be all in on HANA at the cost of the other products customers are running now.

This was my first (and hopefully not last) chance to attend an event like this.  It was intriguing to see how an event like this unfolds outside of the large venues like ASUG Annual Conference/Sapphire and TechEd.  Keep an eye out and see how these announcements play out in the month or so before ASUG Annual Conference/Sapphire- will there be any play in the messages delivered for the conference?  It should be an interesting ride.

To report this post you need to login first.

1 Comment

You must be Logged on to comment or reply to a post.

  1. David Hull

    Great blog, Derek!

    On the HANA testing, you will see a white paper by Sapphire, with all the technical details you care to read.

    In my opinion, some of the most important details are that:

    1. The data was loaded into a partitioned fact table, but no other query speed-up mechanisms were used – no aggregates, indexes, pre-calculations or caching of any kind (and, as a result, no time required to build and populate such structures!). Just standard BW queries for example pulling customer, materials and such by month, quarter or year.
    2. The ad-hoc queries were things that you cannot build materialized views to support – for example, sliding time window comparisons of one month to the same month in a prior quarter or year, and the same with quarterly or yearly comparisons. These really show the power of HANA, in that they cannot easily and quickly be done today with any RDBMS.

    Using SAP Landscape Transformation software to replicate data in real-time from SAP ECC into HANA, without the need to build complex tuning structures, it is easy to see how HANA can enable BW to be a virtually real-time analytics platform for ERP data.

    Cheers,

    David.

    (0) 

Leave a Reply