Explore a conference talk on developing a highly available and scalable multi-tenant eventing backend architecture for serverless products. Dive into the challenges of scaling Knative Eventing backend components and learn about a new reusable, customizable, and pluggable architecture that optimizes resource scheduling. Discover how this innovative Eventing scheduler has been implemented in Knative Eventing to scale KafkaSource's dataplane and ensure high availability of consumers. Gain insights into tuning Kafka consumer settings for increased throughput in production environments and understand how to apply similar design principles to custom resources.