Container or BI Tool?
We just made our ACF Snowflake Insights app available on AWS Market place. If you are using a Snowflake database you should take a look.
Read more about it here: https://acloudfrontier.com/acf-insights-for-snowflake-app/
But why did we actually build a containerised app instead of using a traditional BI tool? It is for sure easier to create a very fancy Tableau or Microsoft PowerBI report and offer it for free? I know QlikSense and Tableau already offer pre-made reports for fancy visualisations of Snowflake credits and usage.
The primary reason? Me! Personally I don’t like BI reports. I did some BI reporting many years ago, and for me it is a pain. As I was the decision maker it was a very fast and easy company decision. :-)
If we skip my personal opinion the more “professional” reasons are:
- We wanted full integration to the offerings of the cloud provider. Initially we are only using AWS SNS for notifications but we have the opportunity to fully integrate to other AWS services in the future. For more advanced use cases it could be Dataiku or AWS Sagemaker for data science.
- Easy distribution platform, using a container instead of a file(s) makes life easier and feasible.
- Implement a cost outlier using the 1.5 x IQR rule for notifications.
- Only use code (Python) so changes can easily be pushed to a new container and published without requiring visual “flueknepperi” for any changes. Click the danish word to get the English translation… :-)
- CI/CD for an easy development and deployment process.
- Easy to branch code for difference solutions and cloud providers, Azure og OCP.
- Easy to develop new solutions based on code instead of a report in a traditional BI tool.
Looking forward to hear your thoughts! I am sure many PowerBI and Tableau developers have a lot of opinions on how to do above :-)
But remember.. flueknepperi — below is what it actually means in danish.
Have a great day!
