AutoMQ: Agent-Managed Diskless Kafka

Kafka is costly to run partly because brokers own the data they serve. Teams must provision disks, maintain replicas, and move partition data as capacity changes. AutoMQ breaks that coupling: a write-ahead log (WAL) protects new writes, retained data lives in shared object storage, and brokers can scale or be replaced without copying partition history. We’ll explain how this diskless design reduces infrastructure and storage-management work, and how Self-Balancing and Auto Scaling respond to changing workloads.

That simpler foundation also changes how the cluster can be managed. Rather than building a complex control plane around broker-local state, AutoMQ Software lets an AI agent use documented interfaces and cluster signals to plan, carry out approved changes, and verify the results. We’ll show the diskless architecture first, then demonstrate the agent managing it through a concrete operational task.

By attending this session, your contact information may be shared with the sponsor for relevant follow up for this event only.