What is RACS?

RACS (Remote Audio Caching Server) is an persistent, append-only audio cache with time-indexed range queries and format conversion. Unlike blob stores and traditional databases, RACS is purpose-built for storing, querying, and streaming audio data at scale.

Key Features

• Format Conversion: On-the-fly transcoding to MP3, Opus, and WAV formats to adapt to client or bandwidth requirements.
• Time-Range Queries: Retrieve precise audio segments by timestamp for efficient slicing, analysis, or playback.
• Metadata Management: Store and query audio-specific metadata such as sample rate, bit depth, and channels.
• Caching & Persistence: Hybrid memory–disk storage with LRU caching for fast access and durable persistence.
• Programmability: Embedded Guile Scheme interpreter for direct, scriptable interaction with the server.

Storage Engine

RACS uses a multi-tiered architecture inspired by LSM (Log-Structured Merge) trees:

1. Memtables: Incoming audio blocks are written in memory and batched for efficient disk flushes.
2. SSTables (Time-Partitioned): When memtables reach capacity, they are flushed as immutable time-partitioned files on disk. This layout enables fast range queries by timestamp.
3. LRU Cache: Frequently accessed SSTables are cached in memory, minimizing disk I/O and improving response times.

This architecture ensures a balance between high write performance, query efficiency, and data durability.

Why Use RACS?

Audio deserves more than generic storage.

RACS is built for systems where audio is dynamic, time-indexed, and performance-critical. It replaces complex pipelines and slow object stores with a single, purpose-built engine for fast, precise, and reliable audio access.

When you need to treat audio as data — not files — RACS is the right tool.

⚠️ MVP Stage

RACS is currently in the MVP (Minimum Viable Product) stage. All the key features mentioned above are functional but still undergoing testing. Planed enhancements and new features will be added in the near future. Feedback is welcome!

Is RACS Open Source?

RACS is distributed under the RACS Source Available License (RACS-SAL-1.0). It is free for personal, educational, and non-commercial use.

Commercial use requires a paid license. For licensing inquiries, contact sales@racslabs.com

Getting Started

The quickest way to run RACS is using docker.

If you do not have docker on your machine, Install Docker before continuing.

    docker build --tag racslabs/racs .

    docker run --network=host -p 6381:6381 racslabs/racs

You can now connect to the server on localhost:6381.

Other Build Options

• Build From Source

Documentation Generation

To generate the docs, run the following commands:

    pip install -U sphinx

    pip install piccolo-theme

    cd docs && make html

Open the index.html file under docs/build/html in the browser.

Note

python 3+ is required for doc generation.

Client Libraries

RACS provides client libraries for multiple languages:

• (Python) py-racs — fully usable
• (Rust) rust-racs — fully usable
• (C) racs-client — usable