So, after 20 years of building applications on massively parallel processing (MPP) shared-nothing database architectures, we thought we'd better get in on this new fangled 'big data' bandwagon!
Apart from being open source, the main difference between Hadoop and the MPP database technologies we know and love (Teradata, Netezza, Greenplum) is that we can now develop scalable business applications using procedural languages such as Java, and not 'just' SQL.
Building a house with nothing but a hammer has become second nature to us over the years, but it'll be nice to have a full box of tools available for a change.
With our real-world project expertise and in-depth MPP architectural knowledge, the adoption of Hadoop into our armoury feels like a very natural fit.
Here's hoping :-)