Analysing EC Builds

[HamzaJ]

I was looking for a way to analyse EC Builds in an easy way to identify where potential bottlenecks within the buildprocess could be. I have created a python script which exports EC build logs (can only export latest build logs) to JSON and converts these into CSVs which can be imported into an ElastiCube.

So I have created a simple dashboard and EC-model. Feel free to adjust to your own needs and perhaps share them so I can learn aswell 🙂 (Do not forget to change the folder in de CSV-connection to the folder where you saved your CSVs)

Download ECmodel + Dashboard: LINK

Download Python: LINK

Configure the settings.yml file with your admin credentials and sisense url. I am not a python-king so adjust freely if you see something thats worth improving.

[KalonWellSky]

The monitoring option under System Management offers dashboards that can be used to monitor CPU, memory and disk space used on your servers.  What I do have this dashboard up on one screen and monitor the CPU during the build run to watch for high CPU usage.  When it spikes, what is going on in the build?  Using this, for example, I found that Sisense does not handle the "Like" query option.  I removed it and replaced it with a list of all possible actual values in a table and the CPU spike was gone and the field built faster.  This dashboard is also useful for determining the impact on memory that changes in your build make.  What is the impact of limiting fields, removing indexes, etc. and when you got the size within your needs.