Pick up a Book… & Throw it Out The Window!

When I learned to play the guitar, the first thing I did was obviously buy a guitar. Then I got a guitar teacher. And then I practised, and then practised, and when I had finished practising I practised some more. I made many mistakes. I got frustrated. But I never gave up. And over time I got better, made less mistakes, picked up some tips and tricks to make playing easier. I developed an ear for notes. I understood scales and keys and relative minors. And now I’m good. not great, but I can play a bunch of Smashing Pumpkins songs and really have a lot of fun playing the guitar without really putting too much effort into practising because I put so much effort in the beginning.

90% of my learning was based on practice. 10% was based on reading books and putting it into practice. The books came some time after buying a guitar. To be clear: I did not learn the pentatonic scale and it’s different position on the fretboard by reading books. I put the theory into practice. To really hammer the point home: I did not learn how to play the guitar by picking up a book and reading it. It was merely a guide to point me in the right direction.

So how does this relate to databases? 

I really don’t get what Hadoop is. Like I understand it is working with large data sets on distributed computers, but that doesn’t doesn’t mean I get it. I’ve read about it, attended talks about Hadoop and watched demos but I just don’t get it. In fact I struggle to understand what is is I don’t get. And I don’t like not getting it.

Then it hit me today; if I learned to play the guitar not by reading books and watching people play the guitar, but by actually playing the guitar, so why don’t I learn about Hadoop by actually working with Hadoop?

To that end, over the Christmas period, I’m going to build a Hadoop cluster using Raspberry Pi’s. Ironically enough, there is a book about it, and there’s also blogs about it, and no doubt I’ll use them as a guide, but if I don’t actually play with Hadoop, I’ll never understand it.

I’m going to post bits of my experiences here, and his post is pretty much a public commitment to ensure that I can’t back out, just like how I can’t back out of that ultra marathon now that my boss, who suggested I run it, has told everyone at work I’m running it with him.

Wish me luck (on both counts)

Author: Richie Lee

Full time computer guy, part time runner. Full time Dad, part time blogger. Pokémon Nut. Writer of fractured sentences. Maker of the best damn macaroni cheese you've ever tasted.

4 thoughts on “Pick up a Book… & Throw it Out The Window!”

  1. Sounds like New Year’s resolution before New Year. ☺ I will keep the fingers crossed.
    By the way I was wondering what was a reason to choose Raspberry. Could you please share book’s name to learn from basics?
    Cheers,
    Krzysztof

    1. Hello!

      The main reason I chose the Raspberry Pi is because I wanted to find a fun way of learning Hadoop. So I found this book and decided to go for it.

      https://www.packtpub.com/hardware-and-creative/raspberry-pi-super-cluster

      With regards to learning the basics with Raspberry Pi, it really depends on your level of knowledge, and what you want to do with it. I’ve never actually used the Raspberry Pi, but I am familiar with computers, but I’ll probably get this guide as a good starting point. It’s written by the people who made the Pi, so it should be decent.

      http://www.amazon.co.uk/Raspberry-User-Guide-Eben-Upton/dp/1118795482

      Do you have any ideas what you want to do, or do you just want to play around?

  2. I’m BI developer using MS stack (DB, SSIS, SSAS and SSRS) for quite big retail project and have many performance issues, specially related to SSAS engine. I am worrying, as many of my colleagues, what is a future for MS BI stack. My personal impression is that MS does not know where to go or how to deliver enterprise level BI solution for the future.
    I am not really sure that PowerBI, DAX, In-Memory and Excel is the answer. That is why I decided to learn something new from basics. I am still not sure which way to choose, maybe Hadoop, maybe Teradata or something similar to PDW… What is your opinion?

    1. Tricky. I find that as worthwhile it is to learn something new, you only really get to grips with it when using it as part of your job, not sure you agree with me on this. But I don’t think that there’s any wrong technology to go install and get experience with.. as long as it’s not Oracle!

      For me, I’m installing and learning Hadoop because I don’t fully understand it and would like to get to understand it better. If you look at this slide from Microsoft, PDW (now Analytics Platform System) actually integrates with Hadoop. There’s also a great “Hadoop 101” article on the same site. So I’d recommend getting to grips with Hadoop just because of it’s popularity.

      http://www.jenunderwood.com/2014/04/18/microsofts-data-culture/
      http://www.jenunderwood.com/2014/10/07/kimball-big-data-warehousing/

      Working in the Microsoft/BI field, I’ll probably end up having to get familiar with Hadoop and “Big Data”, and the Raspberry Pi project looks as fun as anything else I could learn!

      w/r/t to Microsoft and BI: SQL, as a product, is huge. I agree with you, I’m not fully certain where Microsoft are heading with BI, and I thought that maybe the lack on any new features in the BI arena for SQL Server 2014 might be an indication that Microsoft are moving away from SSIS/SSRS/SSAS. But then I went to a public event at the Microsoft offices and Mark Souza commented that for those of us wanting new features in the BI space to wait for SQL 2015.

      It’s worth pointing out the massive fallout PASS (the organisation) had recently when the board (a majority of whom are BI Professionals mind) decided to rename themselves from “Professional Association for SQL Server” to just “PASS”. No acronym. The reason for this being that the original name no was out dated, and no longer accomodated the varying types of data professionals (including Business Analytics users) that work with SQL Server. There’s the PASS BA Conference next year (that covers the BI stack, but is not for BA professionals apparently).

      http://cwebbbi.wordpress.com/2014/09/22/why-i-support-the-pass-name-change-and-its-efforts-to-expand-into-the-ba-community/
      http://cwebbbi.wordpress.com/2012/12/03/the-pass-business-analytics-conference-is-not-the-pass-business-intelligence-conference/

      I think to get a better idea on where Microsoft think they are heading with BI, it’d be worth checking out the BI Track at SQL PASS right now, and the agenda for next years SQL PASS BAC.

      I also am interested to hear about your performance concerns on SSAS.

      Apologies for the wordy reply!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s