Qlikview / Qliksense Environment Architecture

[#1]

Hi all,

I am designing a mixed Qlikview Qliksense Clustered Enviroment for a customer and i need your comments/feedback.

So basically there will be 4 machines .

1. QVServer 1

2. Qvserver 2

3. QVWS,QVPublisher 1, Qliksense Server1

4. QVWS, QVPublisher 2, Qliksense Server2.

QVServer 1 and 2 will be Clustered.

The same applies for QVPublisher 1 and 2 and also Qliksense 1 and 2.

My main concern is storage. The main model size will be 100gb approx.

So my sollution is (OS = operating system, VM = virtual memory)

QVS1. 100gb +OS +VM storage

QVS2. OS +VM storage

QVPub1 --> 100GB+ 100GB for reloads + 75 GB for QVD files. +os +vm storage +100gb for qliksense

QVPub2--> OS +VM storage

Qlikview server 1 will share the storage with QVServer 2

Publisher 1 will do the same with publisher 2

The same applies for Qliksense

Need your comments regarding the architecture in general, as well as the suggested storage architecture.

Furthermore , does any of you have any info to share regarding licenses required to build a Qliksense CLustered Server?

Thank you

[vardhancse]

Hi,

In general not recommended to load such a huge data even after clustered.

Because if we load such a huge data and if the end users are more then it will take so much of time to see/drill down reports and dashboards.

Instead there are many techniques like splitting data/QVD based on some years

or

loop and reduce option so that the front end dashboard will be much easier to reload and drill downs can be done easily.

[#3]

Customer wants entire ungrouped dataset loaded in single file. I am aware of those techniques and best practices and i am preparing demos that i can show him on live environment. My main concern now is the architecture.