OLAP analysis with HDP 2.6

[#1]

Hello,

I'm doing a PoC with HDP 2.6 on small cluster (3 nodes) to evaluate the feasibility of olap analysis for my company. My two options are :

- Kylin 2.0

- Druid (the technical preview in HDP 2.6)

I would like to know if there are some persons doing the same tests and if you could you share your insights ?

I'm not very advanced right now. I'm trying to build the sample cube with Kylin 2.0. It's really the beginning for me.

Thank you in advance for yours answers.

Arnault

[#2 · Accepted solution]

Hi @Arnault Droz i have not used Kylin at All, but i can speak for druid + hive and Superset (visi tool) if you have any questions about that.

Here is a link to blog about some Star Schema Benchmark (6 Billion rows scale setup) This should give you an idea about the performance that you should expect if you go the Druid route to store the cubes.

Please let me know if you have further question.

[#3]

Hello @Slim, Thank you a lot I will check your link.