Hi @yali ,
Any news about porting to azure ?
It is for us a hard requirement to integrate snowplow to our stack.
Thank you
I think @josh might be well placed to update on this one. I think the work is underway for this in terms of thinking how to best structure this on Azure - and the recently released Snowbridge now makes it possible to send events into Azure Event Hubs which makes it possible to run some of the Snowplow pipeline in Azure.
Hi @aka no real updates in this space just yet - we are planning it out and hope to start making real progress in this space this year but likely not to start in earnest for a while yet.
Technically the assets that support Kafka can be used already in Azure as EventHubs has compatibility with the Kafka Consumer/Producer APIs: Use Azure Event Hubs from an Apache Kafka app - Azure Event Hubs | Microsoft Learn
This will get you the Collector & Enrich (mostly) working in Azure but from there we have no loader support BUT as @mike mentioned you could then use Snowbridge (which does support Kafka) to relay Enriched data elsewhere.
This is not a configuration we have tested aggressively ourselves but would be happy to provide assistance if you wanted to go down that route yourself in getting the OS components working with EventHubs via Kafka.
It’s worth mentioning that Snowplow has support for Databricks on Azure. The pipeline components need to run in AWS or GCP, but then the RDB loader can load events cross-cloud into Azure Databricks.
I mention this just in case any other Azure customer finds this thread! If can be a good solution for companies who just need the data in Azure, but who don’t mind running infrastructure in other clouds.
Quick correction - Snowbridge doesn’t support Kafka source at present, so unfortunately that solution isn’t an option at the moment.
Finally! Snowplow is now available on Azure https://snowplow.io/blog/announcing-open-source-azure-support/
In case anyone hasn’t seen - we released open source support for Azure earlier this week