Last weeks analytics stories you may have missed

Let's start with 3 stories on upcoming technologies and finish with one on that ever elusive single version of the truth...

• How Apache Spark Is Transforming Big Data Processing, Development.  Spark is the latest "hot" analytics project husbanded by Apache.    Most articles I've seen so far are either pretty superficial or too technical to help much in my opinion.  This one is a bit more nuanced, describing how Spark is different to other processing frameworks such as MapReduce and how it can fit into the Hadoop ecosystem.


• 9 NoSQL Pioneers Who Modernized Data Management.  This almost seems like a nostalgic look back at some tech greats, but I would guess most of the people here have yet to reach the apogee of their careers.  A pretty good skim if you want a who's-who of the next generation of database technology.


• Larry Ellison's yacht isn't threatened by NoSQL – yet.  I like The Reg.  The articles are usually written in the no-nonsense pithy style that we favour over there.  This one comes with a warning though...  The statement that MongoDB might be in the top three databases soon is just utter tosh.  That conclusion is only possible on the data shown here if you're willing to overlook the log axis on the chart shown, as the author seems to have managed. That said, there are some interesting perspectives in here on what NoSQL databases are good for - or not - especially if you read the comments.  And fair warning, sometimes the Brit comments should come with a PG-13 sticker...


• Finding a Single Version of Truth Within Big Data.  Ah yes.  For all the years that I've been involved with BI and analytics, people have been searching for a single version of the truth.  That is, everyone in a company having access to a single, consistent set of data, showing consistent and irrefutable measures and metrics.  Organizations have wanted it, and vendors have been only too happy to sell them ways to try and achieve it.  Truth be told, few companies probably ever got there, because it's as much a people problem as a technology problem.  This article suggests that big data won't make it easier to achieve that goal as we might have assumed, it will make it harder.  So we should stop wasting our time and figure out how to live with multiple versions of the truth instead.