Liftbridge provides lightweight, fault-tolerant message streams by implementing a durable, replicated, and scalable message log. The vision for Liftbridge is to deliver a "Kafka-lite" solution designed with the Go community first in mind. Unlike Kafka, which is built on the JVM and whose canonical client library is Java (or the C-based librdkafka), Liftbridge and its canonical client, go-liftbridge, are implemented in Go. The ultimate goal of Liftbridge is to provide a lightweight message-streaming solution with a focus on simplicity and usability. Use it as a simpler and lighter alternative to systems like Kafka and Pulsar or to add streaming semantics to an existing NATS deployment.
See the introduction post on Liftbridge and this post for more context and some of the inspiration behind it.
Maintainers
This project is maintained by Basekick Labs, creators of Arc.
Requirements
- Go 1.25.3+
- NATS Server v2.10.0+
Documentation
Performance
Liftbridge delivers high-throughput message ingestion with durable storage. The following benchmarks were run on a single node with replication factor 1:
| Configuration | Throughput | Latency (P99) |
|---|---|---|
| 1 publisher, synchronous | ~30K msgs/sec | 306µs |
| 1 publisher, pub-batch=100 | ~139K msgs/sec | 1.0ms |
| 4 publishers, pub-batch=100 | ~241K msgs/sec | 2.0ms |
| 256 publishers, synchronous | ~200K msgs/sec | 1.4ms |
For comparison, NATS JetStream achieves ~220K msgs/sec with similar settings (pubbatch=100, file storage, RF=1).
Running Benchmarks
# Producer benchmark with async batching (recommended for high throughput) go run ./bench/producer \ --servers localhost:9292 \ --messages 100000 \ --message-size 256 \ --concurrent 4 \ --pub-batch 100 \ --ack-policy leader \ --create-stream # Consumer benchmark go run ./bench/consumer \ --servers localhost:9292 \ --stream bench-stream \ --expected 100000