topic Re: How to reduce application size? in QlikView

How to reduce application size?

_rohitgharat — Sat, 06 Oct 2018 18:07:02 GMT

Hi Folks,

I need your inputs on how to reduce the application size or what is the best optimal way.

1. In my application there are two parts Summary and Detailed reports and we are showing historical data for past 4 years. Due to which the application size has crossed above 100 MB. But we have limitation of 50 MB for each application to deploy it on production.

How can i resolve this problem?.The user requires data for past 4 years in the dashboard.

Is it the best approach to split the dashboard? or what could be another solution for it keeping in mind performance and mainainence.

Please help with your valuable inputs.

Thanks in advance

Regards,

Rohit

Re: How to reduce application size?

captain89 — Sun, 07 Oct 2018 11:05:58 GMT

Hi,

i need more infomation to help you but I'll try to do it anyway.

In my opinion not every detail is necessary in every analysis.

You can make and etl application to extract a summary data. Then you can split different data in different analysis.

For example if you do an application for profitability analysis you need a different cost by item for every year but you don't need the detail for every component of the bottom of material.

In this way you can do a detailed cost analysis with another application and save only one row for profitability and sales analysis (put cost of material, labour cost and process in every invoice without details).

The traditional approach is Etl > qvd > Data Model > App

So you can use one etl to store data in different qvds and make more data models with the same data.

hope that helps

Re: How to reduce application size?

Or — Sun, 07 Oct 2018 11:41:31 GMT

Download QlikView Document Analyzer from Tools | Qlikview Cookbook

Run this for your document and check the results. This should tell you which fields are using up most of your memory. You can then either drop unnecessary fields, or use assorted methods such as hashing, removing timestamps from dates, splitting timestamps from dates (and rounding the timestamp to the nearest second/minute/hour if possible), rounding numbers if you don't need the extra digits, etc. to reduce field sizes.