databricks

Databricks announces General Availability of Real-Time Mode (RTM) for Apache Spark Structured Streaming, enabling millisecond-level latency without leaving the Spark ecosystem. RTM replaces periodic microbatching with continuous data flow through three architectural innovations, achieving latency comparable to Apache Flink in benchmarks while eliminating the need to maintain separate specialized streaming engines. Enabling RTM requires only a single configuration flag on Databricks Runtime 16.4+. Early adopters including Coinbase, DraftKings, and MakeMyTrip report 80%+ latency reductions and sub-100ms P99s for fraud detection, ML feature computation, and personalization workloads.

Announcing General Availability of Real-Time Mode for Apache Spark Structured Streaming on Databricks

Powering industry-leading customers and use cases

Spark RTM: Up to 92% faster than Flink, enabling teams to operate less infrastructure, move faster