Apache Kafka integration

[parano]

Is your feature request related to a problem? Please describe.

Streaming serving is a common pattern for deploying models, where the ML model is applied on streaming data sets.

Describe the solution you'd like

• Docs on an end-to-end model serving solution built with BentoML and Kafka
• Build related API & Integration in BentoML if needed

Describe alternatives you've considered

Welcome suggestions!

Additional context

• Streaming serving with Apache Spark is covered here in feat: Support distributed batch inferencing job on Apache Spark cluster #890
• Thanks, @kaiwaehner for suggesting this. Here's a related post on this topic by Kai https://www.kai-waehner.de/blog/2020/10/27/streaming-machine-learning-kafka-native-model-server-deployment-rpc-embedded-streams/

[awalvie]

Me and @pandafy would like to work on this issue. We're both MLH Fellows.

[parano]

@awalvie @pandafy let me know if you have any questions or would like to discuss your approach & design before starting to implement it. I'm always available in the BentoML community slack and MLH discord.

[Styren]

What is the status on this @awalvie ?

[awalvie]

Hey, I've moved away from the issue at the moment, please unassign me so that others can take it up.

[parano]

Hi @Styren - we are looking to speak with users who are interested in this feature, let me know if you'd be open to sharing a bit more about your use case and requirements.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[martjushev]

This issue is listed in "Done" column in the roadmap. I think it can confuse people into thinking that this is actually implemented