Big Data was The Big Topic at IOD
This year was my first IOD. I hope it won’t be the last because it was a really good conference. These big shows are always difficult, despite the mobile apps, the maps, the schedules, the signage and many helpful show guides. I still have difficulty getting to the sessions I want to see or being frustrated that I want to see two clashing sessions, never mind the miles you walk at these events; I suppose that’s some contribution to exercise. I did manage to get out for a run early one morning - UK jet lag getting me up before dawn every day. I met quite a few people wandering back to their hotels from the night before.
I was working the IBM Netezza booth and presenting the IBM Netezza 101 so mostly I should have been getting questions about the basic IBM Netezza proposition: Speed, Simplicity, Scalability, Smart Analytics and Nice Red Uniforms (I thought I’d have to search in conjunction with Cardinal Ximenes to get that reference to the top of the list, but it was first hit, and I just lost 10 minutes reading the wikipedia article; glad to see the influence of Monty Python on WarHammer).
But in reality the hottest topic i met at IOD was Big Data and the Data Warehouse. I had blogged about it before the show so it wasn’t really a surprise, and my view - that the technology choice (Hadoop, streaming data processing, relational warehouse processing or some combination) is predicated on the use case seems to have resonance. People seemed to be genuinely confused by traditional RDBMS vendors (naming no names) telling them that you just parse your unstructured data and load it into the warehouse with your structured data to do the analytics, and Hadoop vendors telling them you just unload your warehouse into the grid with your unstructured data to do analytics.
This domain really seems to be populated by folk with their own chosen hammers thinking everything is a suitable nail. I prefer the IBM position that you need a full toolkit for all the analytic workloads and the challenge is how you integrate them into a logical data warehouse (what we’ve been calling Smart Consolidation).
Ok, you can say that, to the man with a full toolkit everything looks like a unique requirement, but I think history is going to favour integration of multiple technologies here over coercing all use cases into one technology. I suspect there will be more clarity by next IOD, and i’m reasonably confident I’ll be vindicated.
This week we started the European leg of the IBM Netezza Roadshow with a very a successful opening show in Paris. We’re in London next week, Milan the following week and finishing in Frankfurt on December 1st. The themes will all be IBM Netezza related and there will be lots of experts on hand, technology and industry specialists, to share their knowledge and experience (check out the agendas at the site linked above). So if you have an interest in IBM Netezza, come along.