Skip to main content

RFC 7932 - Brotli Compressed Data Format

Published: July 2016
Status: Informational
Authors: J. Alakuijala (Google Inc.), Z. Szabadka (Google Inc.)

Abstract​

This specification defines a lossless compressed data format that compresses data using a combination of the LZ77 algorithm and Huffman coding, with efficiency comparable to the best currently available general-purpose compression methods.

Status of This Memo​

This document is not an Internet Standards Track specification; it is published for informational purposes.

This document is a product of the Internet Engineering Task Force (IETF). It has been approved for publication by the Internet Engineering Steering Group (IESG). Not all documents approved by the IESG are a candidate for any level of Internet Standard; see Section 2 of RFC 7841.

Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at ````http://www.rfc-editor.org/info/rfc7932\````.

Table of Contents​

    1. Introduction
    • 1.1 Purpose
    • 1.2 Intended Audience
    • 1.3 Scope
    • 1.4 Compliance
    • 1.5 Definitions of Terms and Conventions Used
      • 1.5.1 Packing into Bytes
    1. Compressed Representation Overview
    1. Compressed Representation of Prefix Codes
    • 3.1 Introduction to Prefix Coding
    • 3.2 Use of Prefix Coding in the Brotli Format
    • 3.3 Alphabet Sizes
    • 3.4 Simple Prefix Codes
    • 3.5 Complex Prefix Codes
    1. Encoding of Distances
    1. Encoding of Literal Insertion Lengths and Copy Lengths
    1. Encoding of Block-Switch Commands
    1. Context Modeling
    • 7.1 Context Modes and Context ID Lookup for Literals
    • 7.2 Context ID for Distances
    • 7.3 Encoding of the Context Map
    1. Static Dictionary
    1. Compressed Data Format
    • 9.1 Format of the Stream Header
    • 9.2 Format of the Meta-Block Header
    • 9.3 Format of the Meta-Block Data
    1. Decoding Algorithm
    1. Considerations for Compressor Implementations
    • 11.1 Trivial Compressor
    • 11.2 Aligning Compressed Meta-Blocks to Byte Boundaries
    • 11.3 Creating Self-Contained Parts within the Compressed Data
    1. Security Considerations
    1. IANA Considerations
    1. Informative References

Appendices​

  • Appendix A. Static Dictionary Data
  • Appendix B. List of Word Transformations
  • Appendix C. Computing CRC-32 Check Values
  • Appendix D. Source Code
  • Acknowledgments
  • Authors' Addresses

Overview​

What is Brotli?​

Brotli is a generic-purpose lossless compression algorithm developed by Google in 2013. It is designed to compress data with compression ratios comparable to the best currently available general-purpose compression methods, while providing comparable decompression speeds.

Key Features​

Technical Foundation:

  • LZ77 algorithm variant (sliding window matching)
  • Huffman coding (prefix encoding)
  • Second-order context modeling
  • Static dictionary (predefined common strings)

Performance Characteristics:

  • Compression Ratio: 15-25% better than gzip
  • Decompression Speed: Fast, comparable to gzip
  • Compression Speed: Slower than gzip (adjustable by level)
  • Window Size: 1 KB to 16 MB (configurable)

Typical Applications​

1. HTTP Content Encoding

Accept-Encoding: br, gzip, deflate
Content-Encoding: br

Primary use cases:

  • Static HTML/CSS/JavaScript files
  • Web fonts (WOFF2 uses Brotli internally)
  • API responses

2. Font Compression (WOFF2)

WOFF2 standard:

  • Built-in Brotli compression
  • Font files reduced by 30-50%
  • Supported by all modern browsers

3. Offline Data Compression

Applications:

  • Software package distribution
  • Data archiving
  • Android APK resources

Compression Levels​

Level    Ratio      Speed       Use Case
-----|----------|-----------|------------------
0    | Low      | Very Fast | Real-time (not recommended)
1    | Low      | Fast      | Dynamic content
4    | Medium   | Fast      | Dynamic content (default)
5    | Good     | Medium    | Dynamic content
6    | Good     | Slow      | Static resources
9    | Very Good| Very Slow | Static resources
11   | Best     | Slowest   | Offline pre-compression

Recommendations:

  • Dynamic content: Level 4-5
  • Static pre-compression: Level 11
  • Offline archiving: Level 11

Browser Support​

Browser         | Version | Support
----------------|---------|----------
Chrome          | 50+     | ✓ Full
Firefox         | 44+     | ✓ Full
Safari          | 11+     | ✓ Full
Edge            | 15+     | ✓ Full
Opera           | 36+     | ✓ Full
IE              | All     | ✗ None

Mobile:
iOS Safari 11+: ✓
Android Chrome: ✓
Samsung Internet: ✓

Coverage: >95% of modern browsers

Brotli vs Gzip​

Advantages:

  • ✓ Better compression ratio (15-25% improvement)
  • ✓ Fast or comparable decompression speed
  • ✓ Especially suitable for text content
  • ✓ Static dictionary improves HTML/CSS compression

Disadvantages:

  • ✗ Slower compression speed
  • ✗ Higher CPU usage
  • ✗ Not suitable for real-time high-QPS scenarios
  • ✗ No support in legacy browsers

Migration Strategy:

  1. Static resources: Use Brotli immediately
  2. Dynamic content: Evaluate before deciding
    • High QPS: Keep gzip
    • Low QPS: Can use Brotli
  3. Hybrid strategy:
    • Static: Brotli pre-compression
    • Dynamic: gzip real-time compression
  4. Fallback: Always provide gzip as backup

Summary: Brotli is a modern compression algorithm optimized for the Web, particularly suitable for pre-compressing static content. Through better compression ratios and fast decompression, it significantly improves Web performance and has become a standard component of the modern Web stack.

Last updated on