Welcome

Hyper-Efficient Message Streaming

at Laser Speed

High-performance, persistent message streaming platform written in Rust, capable of processing millions of messages per second with ultra-low latency.

Named after the Italian Greyhound - small yet extremely fast

Sub-ms latencyTail latency under 1ms at P99

Millions msgs/secMulti GB/s throughput on a single node

Zero-copy I/Oio_uring + vectored writes to disk

Getting Started

Install, configure, send your first messages

Architecture

Thread-per-core, io_uring, shared-nothing design

Connectors

Source & sink plugins for data integration

SDKs

Rust, Python, Java, Go, Node.js, C#, C++

Server Config

Tune performance, storage, and security

CLI & Web UI

Manage streams, topics, users from terminal or browser

Connectors: Data Integration Pipeline

Sources

PostgreSQLElasticsearchRandom

Iggy ServerThread-per-core + io_uring

TCPQUICWSHTTP

Sinks

PostgreSQLMongoDBElasticsearchIcebergQuickwit

MCP Server40+ LLM tools Web UIDashboard CLITerminal 7 SDKsAll languages

Why Iggy?

How Iggy compares to traditional message streaming platforms

TraditionalIggy

Runtime

JVM / Go runtime with GC pauses

Native Rust, zero GC, predictable latency

I/O model

epoll + blocking thread pool for disk

io_uring completion-based, kernel does the I/O

Threading

Work-stealing across shared threads

Thread-per-core, CPU-pinned, NUMA-aware

Serialization

Full deserialization on every read

Zero-copy views into raw buffers

Memory

Heap allocations on hot path

Pre-allocated 4 GiB pool, 32 bucket sizes

Binary

JVM + Zookeeper / KRaft + dependencies

Single ~20 MB binary, no dependencies

How Iggy organizes data

Iggy uses a hierarchical model: streams contain topics, topics contain partitions, and partitions store messages in segments on disk. This gives you flexible multi-tenancy, parallel processing, and efficient storage all in one.

Stream Hierarchy

Stream Topic Partition Segment

The append-only log

Every partition is an append-only log. Messages are immutable once written, consumers can read from any offset, and you can replay the entire history at any time.

Append-Only Log

Messages are appended sequentially. Consumers track their position independently via offsets.

write

consumer

Click any message block to move the consumer pointer. New messages appear from the right.

Key capabilities

Capability	Details
Transport	TCP, QUIC, WebSocket, HTTP with custom binary protocol
Performance	Thread-per-core + io_uring, zero-copy deserialization, custom 4 GiB memory pool
Security	TLS on all transports, Argon2id hashing, AES-256-GCM encryption, granular RBAC with per-stream/per-topic permissions
Connectors	Source & sink plugins for PostgreSQL, MongoDB, Elasticsearch, Apache Iceberg, Quickwit with data transforms
AI Integration	MCP server with 40+ tools for LLM-driven message streaming management
Management	Web UI dashboard (embedded or standalone), CLI with shell completions, Prometheus metrics, OpenTelemetry
Clustering	Viewstamped Replication (VSR) consensus protocol (in development)
Deployment	Single binary, Docker & Helm, NUMA-aware CPU affinity

Ecosystem

Iggy is more than just a server. The project includes a full ecosystem of tools:

Connectors Runtime - dynamically loaded Rust plugins for data integration. Ingest from PostgreSQL, Elasticsearch into Iggy, or forward to MongoDB, Elasticsearch, Apache Iceberg, Quickwit. Built-in data transforms and Prometheus metrics.
MCP Server - Model Context Protocol server exposing 40+ tools for LLM integration. Works with Claude Desktop via stdio and HTTP transports.
Web UI - SvelteKit dashboard for stream/topic management, message browsing with JSON/string/XML decoders, user management, and real-time terminal.
CLI - full-featured command-line interface with named connection contexts, session-based login, and shell completions.
SDKs - client libraries for 7 languages (Rust, Python, Java, Go, Node.js, C#, C++) with examples for each.
Benchmarking - built-in iggy-bench tool with a Yew/WebAssembly dashboard for performance testing.