Inside EMC, our awesome (you really are, Patricia!) Americas/EMEA CTO Patricia Florissi along with the Global Presales team have been running this “Big Idea” series. We did this to open a LOT of people’s minds to what we are seeing.
It’s been posted internally (EMCers and EMC Partners) to EduTube (http://edutube.emc.com) which is a great site where we post stuff (sometimes not for “open to all” distribution), but these are TOO GOOD not to share with the world.
We’re finding that in addition to the more traditional enterprise architectures (which are moving to virtualized cloud models), there’s new sets of customer needs around new kinds of problems. Yes, the word “Big Data” is applied to the problems as shorthand, but they are associated with a basic principle:
Big Data problems are ones who’s inherent scale causes traditional architectural models to break down.
This series started with a grounding in “What is Big Data” post here. It’s great content (one of the most popular videos I’ve ever posted.
The second video broadens the scope to “Understanding cluster architectures” which is a grounding on the design principles that go into these “data locality and parallelization” models using commodity components, planning for failure and other principles. As Patricia noted, “you won’t understand Hadoop without understanding these cluster architecture fundamentals”. You can watch that here.
This most recent video looks into one of the most important examples of a clustered architecture in wide use today – Hadoop. While of course EMC is heavily invested here (Greenplum HD is a leading Enterprise Hadoop distribution, we contribute to the open source Hadoop team, and we are building a 1000 node Hadoop cluster open for people to use), this is useful to ANYONE interested in Hadoop. Check it out!
Feedback as always welcome!!! For EMCers and Partners – check out EduTube!!!
Comments