Ross ROSS = Recommend OSS · open-source software intelligence for agents

nisshi-io/nisshi

Apache Kafka® compatible broker with S3, PostgreSQL, SQLite, Apache Iceberg and Delta Lake observed · 2026-08-28

github.com/nisshi-io/nisshi · homepage · Rust · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

88/100

  • Activity 99
  • Release rhythm 95
  • Longevity 53
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-02. Adoption (stars, forks) is never an input.

  • gap_med: 5
  • age_days: 742
  • days_rel: 33
  • days_push: 7
  • n_releases_24m: 28

Full methodology

Adoption not part of the score

1851 stars · 76 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Nisshi is a stateless, Kafka API-compatible message broker written in async Rust with pluggable storage engines (PostgreSQL, SQLite/libSQL, S3, or memory). It includes an embedded schema registry (Avro, JSON Schema, Protobuf) and can write schema-backed topics as Apache Iceberg or Delta Lake tables.

Use cases

  • run a kafka-compatible message broker without zookeeper or raft consensus
  • store kafka topic data in s3 or postgresql instead of replicated broker storage
  • validate avro, json, or protobuf messages against schemas at produce time
  • stream kafka topics into apache iceberg or delta lake tables
  • spin up an ephemeral in-memory kafka broker for testing
  • migrate off apache kafka without rewriting kafka clients

When to choose

  • you want kafka API compatibility with cheaper, simpler durability via S3 or PostgreSQL
  • you need schema validation and a schema registry built into the broker
  • you want topic data landing directly in Iceberg or Delta Lake for analytics
  • you prefer a single statically linked binary with no cluster coordination overhead

When to avoid

  • you need battle-tested, large-scale Kafka ecosystem features like Kafka Connect, Streams, or tiered storage parity
  • you require broker-side replication semantics and in-sync replicas for strict durability guarantees
  • your tooling depends on Kafka-specific protocols beyond the core client API
  • you need a mature project - it is relatively young with a small download base

Facets

service · maturity active

message-queue streaming serialization database object-storage cli proxy big-data databases developer-tools self-hosted rust self-hosted cloud kafka-compatible apache-iceberg delta-lake schema-registry avro protobuf s3-storage stateless-broker parquet apache-arrow messaging data-engineering linux macos docker

10 sources

Member repositories

RepositoryRoleHealth v2
nisshi-io/nisshimain88

For agents

markdown · JSON · MCP: product_card(name="nisshi-io/nisshi")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem