timeplus-io/proton

The Fastest Unified Streaming SQL Engine in a Single C++ Binary. ⚡ Millisecond latency. 100+ GB/s throughput. Continuously compute real-time context from streams, logs, metrics, events, and CDC.

What it solves

Timeplus Proton is a high-performance SQL pipeline engine designed for real-time stream processing, analytics, and observability. It serves as a lightweight, C++ based alternative to complex frameworks like Apache Flink or ksqlDB, eliminating the need for a JVM or ZooKeeper while providing the ability to both process streaming data and store queryable materialized views for analytics dashboards.

How it works

Proton is built on the ClickHouse engine, extending its storage and computing capabilities with stream processing. It uses a single C++ binary with SIMD optimizations for high throughput and low latency. Users interact with the system primarily via SQL to create external streams (e.g., from Kafka or Redpanda) and external tables, using materialized views to transform and route data in real-time.

Who it’s for

It is intended for data engineers and developers building real-time ETL pipelines, telemetry pipelines with alerting, and real-time feature pipelines for AI/ML models that require low-latency, high-throughput processing of live data.

Highlights

  • High Performance: Capable of delivering up to 90 million events per second (EPS) with millisecond end-to-end latency.
  • Zero Dependencies: Distributed as a single binary (<500MB) without requiring a JVM or ZooKeeper.
  • Broad Connectivity: Native support for sources and sinks including Kafka, ClickHouse, MySQL, Postgres, MongoDB, and S3/Iceberg.
  • Streaming SQL: Supports tumble, hop, and session windows, watermarks, and incremental materialized views.
  • Extensibility: Supports User Defined Functions (UDFs) in Python and JavaScript.

Related

  • Project
    arkflow-rs/arkflow

    High‑performance Rust stream processing engine that integrates AI capabilities for real‑time data processing, inference, anomaly detection, and complex event processing.

  • Project
    quixio/quix-streams

    An open-source Python framework for real-time data engineering and machine learning on Apache Kafka streams, providing a tabular DataFrame API for reliable ETL pipelines.

  • Project
    airbnb/chronon

    A data platform for AI/ML that abstracts data computation and serving, ensuring point-in-time consistency between offline training data and online real-time feature serving.

  • Project
    hosseinmoein/DataFrame

    A high-performance C++ library for in-memory tabular data analysis and transformation, providing a fast alternative to Pandas and Polars for large-scale statistical and ML datasets.

  • Project
    Netflix/metaflow

    A human-centric framework for building and manage real-life AI and ML systems, streamlining the transition from prototyping to production deployment.