So as you may have learnt, like London Buses, HDEs don’t come for a while… and then they come in 3s! And so I’m very please to welcome Aron MacDonald, who works for Shell in London. I suggest you check out Aron’s content, because he knows more about HANA and Hadoop than anyone else.
Please welcome Aron to the HDE program. If you’re wondering about the Womble reference, then click on.
Tell us a little around your background in the industry
At University I specialized in Database design & SQL, so was excited when I joined a Major consultancy expecting to work on some leading implementations of Oracle. You can imagine how thrilled I was to then be asked to support a dusty old mainframe version of the Ingres database instead. It wasn’t for long though and I was next thrown into the deep end of SAP implementations. 17 years later, after bouncing around between ABAP, FI/CO & BW I’ve landed back where I started, doing SQL again, and couldn’t be happier.
How did you get into the SAP HANA space?
I was fortunate to have tiny bit of exposure to HANA back in it’s ramp up days, which was enough to rekindled my old interest. The most significant influence though was when SAP made HANA available for developers on AWS in 2012. WOW!!!!
Never before was it so easy to have your own personal development system with SAP’s latest technology. I was then able to build my own HANA PoC’s and present them back to Clients. As an independent contractor this is something that previously would have been almost impossible.
From there my interest quickly spread to integrating HANA with other tools which could be easily launched on AWS. I initially tinkered with Sybase Unwired Platform for deploying iOS Native and HTML5 Mobile applications, but was quickly drawn to the new open source technology of HADOOP. I’m a data person at heart, and the idea of combining these incredibly powerful tools inspires me to learn and explore.
What advice would you give to people looking to learn SAP HANA?
Whether it is the HANA Cloud platform, AWS, Cloudshare etc, then get yourself onto a HANA development system and start building your own PoC. Take something you’re already familiar with and try to recreate it on HANA. Along the way that will drive you to read more and learn more with every little problem you face. As you hit problems though, please search for a solution first before posting questions, in most cases the answer is only a short Google search away.
If you solve something new then please share it, we’re all standing on the shoulders of geniuses and may not have even come close to a solution without their combined knowledge.
Can you tell us a little about the projects you are working on right now?
I’m currently an independent consultant at a major oil company, working on a large HANA side-car solution (8Tb Memory, 16 hosts). There is never a dull moment with an endless supply of interesting topics to explore.
In my spare time I like to setup HADOOP Clusters and test the integration potentials with HANA.
Tell us about one of your HANA war wounds!
Working on version 1 of any software always presents challenges. I’ve been peppered with many number of wounds, but far less than I would have expected. Even in its infancy HANA is a remarkable product.
For me though integration of HANA with other products still proves the single largest pain. Whether it’s DXC, SLT, Data Services or BO, they each seem to have some very sharp edges, which need to be handled with care. They work well most of the time, but as data grows and business requirements get more complex then things start to fail unexpectedly and each needs to traced and fixed in very different way. I do sometimes miss the old days of simply doing an ABAP debug to get to root cause of a problem.
What do you see in the future of HANA?
Being and old school R/3 person I’m perhaps biased, but for me I hope the futures is the shift back of many operational reports to ECC on HANA. BW is not going away, as staging and transforming data is always going to be necessary, but I think it will be done much more selectively. Reducing data staging and reporting directly off of the source tables always seems to make sense to me, assuming the system can cope.
In a much broader sense though, I’d like to see a much tighter integration between HANA and HADOOP, to provide the core technical data architecture of large Enterprises. In-memory computing is definitely the future but so too is the inevitable explosion of data. Data volumes are growing faster than memory prices are falling so Enterprises will need a solution that meets both challenges, otherwise they risk falling behind.
If there was one change you could make to HANA, what would it be?
That’s a hard one as I have a long list, but I’ll cut it down to two.
For HANA standalone it would be the ability to push down ‘Where’ restrictions made on SQL calculation views deep into the underlying logic, without resorting to variables, which can’t easily be used by many external reporting tools.
For ECC on HANA (or HANA side-car) it would be for better native handling of field formatting issues (including leading zeroes and similar meaning columns of different lengths) at the database level (rather than application level). When you are using the wide variety of table and columns designs across ECC, you very quickly hit problems, especially when joining data across modules. In some instances these can be worked around in SLT by adding custom fields, but for ECC on HANA there will either need to be some solid remodeling of core ECC tables (unlikely) or better HANA tools for handling these legacy table design issues, otherwise performance and data presentation may suffer.
Tell us a bit about Aron outside of HANA and work
With 2 small kids under 6, free time is at an absolute premium these days, and those glorious days with freedom to travel with my wife seem like a life time ago. I wouldn’t change things for the world though. I’m originally from Australia, my Wife is from Singapore and our kids are British so trips to visit relatives are always a major affair.
Most weekends we can be found either chauffeuring the kids to their various activities, walking or riding our bikes in Wimbledon Common or chilling out with a coffee in a local café.
I enjoy swimming and going to the beach but have had little time in the past few years to do either, but one day I will again. I’m just waiting for that perfect HANA job role on a tropical island to come up.