Recap: NoSQL East Conference 2009

Last week I had the pleasure of attending NoSQL East 2009, a new conference based around the non-relational data stores that I briefly covered in this post. In short, it was a rather intimate and highly technical conference on the grand scheme of things. There were roughly 120 people in attendance for the two-day event that gathered 16 speakers who discussed, more or less, which NoSQL technology they employed at their company and how they used it.

NoSQL East 2009 - Drinks at TAP
Pre-conference drinks at TAP
NoSQL East 2009 - Auditorium at GTRI
Held in the same auditorium as Ignite ATL 2009.

NoSQL East was unlike any other conference I've been to in how focused it was. Everyone in attendance seemed to be very well-educated and all in the same field, making for a great group of people to chat with. Jon Moore, Ph.D. Computer Science, put it best in his post covering this conference:

There was a telling moment in the first non-keynote talk where the speaker asked the audience "How many people have read the Dynamo paper?" and easily 95% of the audience put their hands up.
John Willis talking about Big Data
John Willis talking about Big Data

The intent of the conference was, as described by one speaker, to be less deep, technical talks and more of the applications and uses of certain types of NoSQL. However, a few speakers did get a bit technical.

Geir Magnusson of Gilt explains vector clocks used by Voldemort
Geir Magnusson of Gilt explains vector clocks used by Voldemort
Gilt's Service Architecture

That being said I rather enjoyed Mike Miller's talk about CouchDB in particular because he went over several case studies and showed how various companies were using and fitting CouchDB into their architecture.

Mike Miller of Cloudant going through some CouchDB case studies
Mike Miller of Cloudant going through some CouchDB case studies

As for most intriguing NoSQL technologies, I liked Kevin Weil's presentation on Pig, a new-to-me higher level language version of Hadoop. Why's that? Because Pig "runs on a cluster computing architecture, on top of which sit several layers of abstraction that ultimately bring the power of parallel computing into the hands of ordinary users." And to show this off, Weil displayed sample Pig code to run through large datasets — of which the equivalent code in another language would have involved many more lines of code.

Kevin Weil of Twitter showing off Pig
Kevin Weil of Twitter showing off Pig (slides)
Silverpop's MapReduce flow
Silverpop's MapReduce flow

Overall

NoSQL East 2009 was a great kick-off to what will hopefully be an annual conference devoted to next generation non-relational data stores. As I mentioned in my last NoSQL post, many people have debated about whether NoSQL is an appropriate term. After this conference it was clear that all of the speakers agreed the best solution is to have smaller RDBMSes where necessary and reserve the heavy lifting and Big Data to these so-called NoSQL technologies. Moore suggested something I rather liked; calling it NOSQL for Not Only SQL.

Alright so you're probably wondering where you can find the intensely technical summaries of this conference. Since I am far from a DBA, I did not attempt summarizing some of the great talks I heard. Instead I'll point you to Eivind Uggedal, who came all the way from Norway for NoSQL East 2009 and groks this stuff better than myself:

As usual, more pictures can be found on my Flickr account.

For those of you that were at NoSQL East 09, what did you think of it? For those of you at home, does this NoSQL talk interest you? I just wish I had a project to make sure of something like Pig or CouchDB. I'm still using my cozy ORM.

Disclosure: I attended NoSQL East 2009 free-of-charge on behalf of the organizers.

Like it? Tweet it.

"Recap: NoSQL East Conference 2009" by @Stammy