The serverless architecture to build a cross cloud supported application

We are working on a billing solution using ELK stack. We have used aws managed ES services for indexing and had Logstash running from a different EC2 instance. The data sources being used are the aws DBR reports, billing APIs etc.

Now we need to scale the same solution to support other cloud platforms (MS Azure & Google Cloud to start with). Each cloud platform has different ways of managing their billing data and have probably different concepts for each services they provide.

What would be the best architecture to build such an application (billing solution) where we can just configure different providers and rest will be take care of by the framework ?

I guess the Serverless framework abstracts these details in some way and have implemented the multi-provider support. Please share some thoughts or point me to some docs available?

Hi @deepu.sundar, at the moment at serverless we don’t support multi-provider out of the box, so you would have to implement the data gathering on each provider yourself and then forward that to an AWS based system. Best way to do it would probably be to have some small pieces running on Google, Azure, … and upload those into an S3 bucket. From that S3 bucket you can then run lambdas to work through that data.