My First

HACKATHON!

I loaded my first Hadoop Cluster on MS Azure and wrote my first Hive query.  

This has been painful, but very, Very worthwhile.  The reason is because Microsoft has a team of experts in various products and this is really all about an intensive training session.  I started out with a group of four guys and one other woman.  About two hours into the "hackathon" proper (as opposed to the leading presentations) three of the guys were gone.  I think at least two of them were here for free food and had no experience with query language, Excel, Dashboard development, databases or their administration.  I'm not sure they could even write code.  The third guy was migrating from policy into coding and wasn't really familiar with query language, Excel, visualization concepts, or analytics, much less the other - basic - items.  

I *intend* to go tomorrow morning.  I left at 8:30pm, an hour before the Microsoft people *could* leave, although a lot of people were wanting to stay the night.  And when I say "a lot," I'd say that out of the original 300 which registered and showed (full room in the morning), there was less than 1/3 left when I left and some of the "kids" were planning on staying up all night.  Microsoft had planned to allow the all-nighter.  I had a conversation with one of the youngsters about, "Why do all the older people leave instead of staying all night?"  

Been there.

Done that.

The feckin' t-shirt has been worn to threads.  I want a bourbon.

But this was a good working experience, but shit, it's like starting up with SSIS & SSRS again.  My eyes are crossed and I feel like an idiot, but if / when I sit down to focus on this again, maybe - just maybe - I'll 'member some of this activity.   I just don't get why MS products are so "complicated" to figure out.  I had to load like twelve new softwares just to get the interface going, but gotta say...  I'm proud of the .csv file I loaded up.

Yep.  That's "Hadoop" - raw files, whatever format you want.  oh joy.  It's a big ole file folder and you specify the parsing routine in the select statement.  

But the thirty day trial versions / access...  I'm not gonna maintain that shit.  So, I guess I'll be dropping back by Hortonworks or somewhere else to spin up some clusters b/c Hive is Apache Hive and so the query language is unchanged.  I go open source b/c I am not gonna pay the corporate $$ but I do need to have access to the technology so I can add to the skillset.