# 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.

Repository: https://github.com/timeplus-io/proton
Canonical: https://ross.abutalabs.com/products/timeplus-io-proton
Homepage: https://timeplus.com
Language: C++
License: Apache-2.0
License Family: permissive
Topics: clickhouse, kakfa, sql, redpanda, single-binary, stream-processing, cpp, iceberg, etl, flink, ksqldb, observability, pipeline, real-time, agentic-ai, context
Last push: 2026-08-13T01:48:19+00:00

## Health v2 (maintenance only)
Score: 93/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 96, longevity 79
- inputs: {"age_days": 1115, "days_push": 21, "days_rel": 31, "gap_med": 11.0, "n_releases_24m": 45}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2247, forks 110 (observed 2026-08-28T04:06:30.171955+00:00)

## What it is
Timeplus Proton is a unified streaming SQL engine shipped as a single dependency-free C++ binary, built on the ClickHouse engine. It ingests from sources like Kafka, ClickHouse, Postgres, and S3/Iceberg, and continuously computes real-time analytics via streaming SQL, materialized views, joins, and alerting.

## Use cases
- run streaming SQL queries over Kafka topics in real time
- build real-time analytics dashboards with millisecond latency
- process CDC events from databases into materialized views
- monitor logs and metrics with continuous SQL pipelines
- replace Apache Flink or ksqlDB without JVM overhead
- create real-time context feeds for AI agents
- ingest and transform streaming data into ClickHouse or Iceberg

## When to choose
- you need high-throughput stream processing with a single zero-dependency binary
- you want SQL-first streaming ETL instead of writing Flink or Kafka Streams code
- you need millisecond-latency analytics on event streams, logs, or CDC data
- you want native connectors to Kafka, ClickHouse, Postgres, and Iceberg in one engine

## When to avoid
- you need batch-only analytics with no streaming requirements
- you require a JVM ecosystem with Flink's extensive connector and state backend options
- you need a fully managed distributed cluster with built-in coordination out of the box

## Facets
- artifact type: application
- maturity: active
- function: streaming, etl, search-engine, monitoring, alerting, database
- domain: analytics, big-data, monitoring, developer-tools
- platform: self-hosted, cpp, cli
- tags: stream-processing, streaming-sql, clickhouse, kafka, materialized-views, cdc, single-binary, flink-alternative, ksqldb-alternative, real-time-analytics, sql, data-engineering, real-time, linux, macos, docker

## Member repositories
- timeplus-io/proton (main) score 93

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:30.171955+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:43:58.096966+00:00, confidence not recorded.
  - readme: https://github.com/timeplus-io/proton (fetched 2026-08-28T04:06:30.171955+00:00, sha 6645658b382f)
  - homepage: https://timeplus.com (fetched 2026-08-29T10:24:14.540084+00:00, sha 860630a6ff5e)
- Data as of 2026-08-30T08:39:29.467469+00:00.