Turning Your Hadoop Cluster Into a Scale-Out BI Server
OK, I am willing to admit it: I am an old-school BI guy, raised in the age of OLAP (do you remember "The OLAP Report" by Nigel Pendse?). My first intro to this world was reading Ralph Kimball's "Data Warehouse Toolkit", which I tore through like it was the best book I'd ever read. I loved the idea that messy data sets in the form of 10s or 100s of tables could be transformed into an optimized format for data consumption - fast, and with concepts that made sense to business users.
Are you like me? Do you get excited by the idea of delivering business intelligence at Hadoop-scale? (if so, well, sorry... and join the club!) Do you love SQL Server Analysis Services (except for the fact that it can't perform at "Big Data Scale"). If you said "Yes!!!!", then what we are doing at AtScale might be relevant. You can learn more about our approach to Scale-Out BI here: http://blog.atscale.com/tech-talk-scale-out-business-intelligence-with-hadoop.
While OLAP as a technology may be dated, the concept is still extremely relevant. This is why AtScale has developed a software platform that delivers on all the things that I (and maybe you?) love about OLAP - interactive performance and a business friendly interface - while doing it at "Big Data Scale". Want to learn more? Drop me a message!
Principal Software Architect- AI , HPC Data Analytics and Hybrid Cloud
8yHello Josh Klahr, Great and thanks for sharing information on OLAP-Cube/Dimension which is my thrust area on Hadoop Ecosystem. No doubt , Ralph Kimball's OLAP toolkit found to be best book written for BI Specialist on conventional RDBMS vertical but when it comes to Hadoop the best of OLAP can be found at AtScale- seems most intelligent OLAP for Hadoop.
All things Data | Chief Problem Solver | Hired Gun
9yI think you're selling out your cluster if you only consider it for BI. I mean it may make sense for you to consider using your cluster to handle the Enterprise's BI responsibility because that's where the data resides. Mixing and matching BI and Analytics work loads may require a rethink on your cluster design, including moving to a compute/storage cluster(s) model. It sounds like AtScale's product mix may be hitting the market at the right time as more traditional companies migrate to 'Big Data'.