AtScale Blog

The BB-8 Reasons You Should Care About BI-on-Hadoop

Posted by Bruno Aziza on Mar 19, 2016
Find me on:

The BI on Hadoop market is at an interesting intersection...even event-wise.  This past week, the Gartner Business Intelligence Summit gathered over 2,000 Analytics enthusiasts, and next week the O'Reilly Strata Conference will see 4,000 Big Data nerds like me discuss the future of Hadoop.  The space is evolving fast.  It's possible you might have missed a few things.  Below a "re-cap" of the Gartner BI Summit and a "pre-cap" of the O'Reilly Strata San Jose event.

1) Don't be L-8 to your BB-8

If you missed the opportunity to win a BB-8 at our Gartner booth, be sure to find us at Strata! BB-8 is a loyal Star Wars astromech.  He's Fast.  He's Agile. He's Intelligent.  And, Everyone Wants to Deploy It.  Just like AtScale.  

Come claim yours at our Strata Booth (#1622).  AtScale's booth is right next to Tableau's: you won't be able to miss us.  Alternatively, if you want to schedule a meeting, book us here.

2) KO on CDOs? Don't Think So.

Gartner's Valerie Logan did an outstanding job outlining the issues Chief Data Officers (CDOs) or Chief Analytics Officers (CAOs) are faced with.  According to her research, the world now counts about 950 CDOs.  Their #1 Job?  Create an information strategy that encompasses information governance and defends the organization against risk.  We think there is more to this, particularly on the "opportunity" side of data.  Listen to my quick perspective here.

3) Spark, Schmark?

Spark has been on fire since it launched.  AtScale has been a key Databricks' partner and together, we have worked tirelessly to make Spark perform in the Enterprise.  Many of our customers use Spark in production today! So, if you are considering using Spark, Impala and/or Hive, check out the latest BI-on-Hadoop benchmark test result report here.  Gartner's Nick Heudecker did a great job explaining what Spark is, and what it isn't.  Do you know?  If you need a refresh, read this.

4) The Sqoop on Hadoop

When AtScale's CEO was Cloudera's largest customer, processing 12B+ signals/day at Klout, very few people even knew how to spell H-a-d-o-o-p!  Those days are long gone. The majority of attendees either have clusters in production or are onboarding Hadoop for Business Intelligence workloads today.  If you're unsure about what you should do next or if you simply want to benchmark your Hadoop maturity against your industry peers,  take the test to receive the results of the Hadoop survey here.

5) On Q for the Gartner MQ?

Josh Parenteau's session on the Magic Quadrant (MQ) was probably one of the most popular ones.  Every customer and every vendors wants to understand Who's Who and Who Isn't.  For years, the BI community has been complaining that the Quadrant was too crowded. Understandably so: the document assessed 24 vendors last year, 9 of which were labelled as "Leaders".  Gartner acted upon this feedback this year and made some drastic changes.  Get the skinny here.

6) Bring Your Own Business Intelligence (a.k.a. BYO-BI)

Between commercial BI solutions, open-source and custom-built analytical tools, the Business Intelligence market is reaching near saturation.  Gartner tracks 20+ BI tools, Forrester says it catalogued 50+ vendors in its BI Vendor Landscape.  Boris Evelson, Vice President and Principal Analyst at Forrester Research notes in his blog last week: "Enterprises, which a couple of years ago had 10+ different BI platforms, today usually only deploy somewhere between 3 and 5".  Bottom line:  Enterprises are trying to consolidate their BI tool ecosystem.  They are not adding new tools, particularly those who are solely dedicated to Hadoop.  For the last 10 years, business users have brought their own tools to the mix.  It is now time to consolidate to the ones they love, for all data sources.  And, there are 2 things that you can be sure of: 1) the average enterprise will always have more than 2 BI interfaces 2) One of them is Microsoft Excel.  Why does this matter? Most BI tools talk SQL, Excel talks MDX.  AtScale is the ONLY BI on Hadoop platform that can talk MDX and SQL.  

7) Go Ahead:  Cheat on Your Data!

Executives often assume that "doing BI on Hadoop" means hiring a new team of admistrators, data-scientists and analysts.  That's scary.  Both for managers and employees.  It's also NOT true.  We joke that doing BI on Hadoop is often about marrying "something old" and "something new".  The "new" is Hadoop.  The "old" is BI.  You can do it all with the team you currently have in place.  Don't listen to the hype or the vendors that want to scare you into buying their services! Start by reading this simple Cheat-Sheet for BI-on-Hadoop.  It's not as complicated as many want you to believe.

8) Your BI-On-Hadoop Journey

It's difficult to start a journey without a map.  Our favorite Gartner BI sessions this year were Mark Beyer's "Big Data Warehouse Stop Panicking" presentation and Svetlana Sicular's "Upgrade Your EDW with Hadoop" session.  Both are extremely experienced analysts and their comments echo the testimonials we've documented in our recent webinars, namely:

  • Embarking on the BI on Hadoop Journey with Yellow Pages (More here)
  • Upgrading Your EDW for Hadoop with WarGaming (More here)
  • Driving Hadoop Adoption with Macy's (More here)
  • The Do's and Don'ts of BI on Hadoop (More here)

We hope that these resources will be useful to you!  In case you're looking for more, be sure to visit the BI on Hadoop Resource Portal here.

See you all next week at Strata San Jose...and Analytically Yours!

 

Topics: Hadoop, Tableau, Gartner, forrester, GartnerBI, Microsoft Excel

Learn about BI & Hadoop

The AtScale Blog is the one-stop shop for cutting edge news and insights about BI on Hadoop and all things AtScale.

Subscribe to Email Updates