Redpanda SQL demo: Analyzing cell tower packet loss

Introducing Redpanda SQL: Redpanda’s new real-time SQL query engine. It is PostgresSQL-based, allowing users to use existing Postgres clients and SQL syntax. General purpose queries can be run on data in motion, and on data at rest in the Apache Iceberg format. RP SQL turns all your Redpanda topics into Postgres tables without moving data, and sits directly inside your BYOC clusters. In this demo, we use Redpanda SQL to examine packet loss across a network cell towers in a mobile phone network. Here, our cell towers are streaming ‘valid’ events. The format is right, the schema validates, but our rules-based observability failed us. Some towers are silently failing with packet loss and other undetected issues. A customer reports packet loss of over 15%, so we need to quickly get to the bottom of it, with a mix of ad hoc & agentic analytics. What you'll see in the demo: Redpanda SQL deployed within an existing Redpanda cluster in Redpanda BYOC Redpanda SQL doing ad hoc analytics on device telemetry in topics using Avro schemas SQL queries that transparently extend beyond a stream’s retention, into its Iceberg ‘history’ An agent using SQL iteratively to explore the data and discover deeper equipment issues Humans and agents working collaboratively to find and validate data issues spanning streams and a lakehouse, then dispatching technicians to fix real problems that require servicing of high value equipment