Best Practices for Quick Sight CI/CD Pipeline Implementation with Python

Q&A
,
1

HEY ALL
I’m looking to implement a robust CI/CD pipeline for Amazon Quick Sight using Python. My environment includes multiple stages (staging, test, and production), and I’m seeking guidance on:

• The most efficient approach to automate Quick Sight deployments
• Best practices for managing assets across different environments
• Recommended tools and frameworks for Quick Sight CI/CD
• Version control strategies for Quick Sight resources

Any insights from those who have implemented similar solutions would be greatly appreciated!

Thanks in advance!

2

Hello Kolodnydan,

I thinks this is what you are looking for. please refer it for better understanding -https://www.youtube.com/watch?v=nWFXVgTnrFg

Hope this helps.
Cheers,
Deep

1 Like
3

Hi @kolodnydan , unfortunately an approach that works for all does not exists. You should take into account many factors, try to answer these questions:

  • what environment specific changes you need to apply before deployment (in addition to changing data sources)
  • how are you going to store assets
  • how are you going to review/apply changes
  • how are you going to implement versioning
  • how are you going to manage access to assets
  • …
    Every answer will affect you solution. You will likely have to chouse from using:
  • CloudFormation
  • Quick Sight API Asset Bundle Export/Import
  • Quick Sight API Describe/Create

Every approach allows to backup/restore assets. It’s just a piece of cake. But every approach has some restrictions related to making environment related adjustments. Making code that allows to apply these adjustments will require some efforts. Correct choice can minimize these efforts. It’s great when your approach allows mostly use parameters in API calls instead of modifying files or objects, so pay attention to this when you read documentation.

1 Like
4

Thanks very helpfull !

1 Like
5

Its actually a Challenge that a lot of Teams are facing. Quicksight Asset Management in its current state can be very complicated and a lot of hours can be spent on getting it to work properly at scale!

For this reason, we are just about to launch a central Platform that makes the entire process as easy and straightforward as it possibly could be!

Feel free to checkout sightflow.io :waving_hand: