Liquibook is a blazing fast C++ limit order book matching engine that provides low-level components for order matching and aggregate depth tracking.

Liquibook is a framework which supports the implementation of a simple order book - the part of a matching engine focused on a single trading instrument.

Visit our Github for:


  • It works with your design choices and any protocol
  • Low-level components for order matching and aggregate depth tracking
  • Memory-efficiency: minimal copying of data to internal structures
  • Speed: between 2.0 million and 2.5 million inserts per second

Benefits of using Liquibook

  • Liquibook is free open source software (FOSS), which means no vendor lock-in and maximum user control
  • There are no license fees whatsoever, and the full source code is available at no cost
  • Optional aggregate depth tracking to any number of levels (static) or BBO only
  • Works with smart or regular pointers
  • Preserves your order model, requiring only trivial interface
  • Preserves your identifiers for securities, accounts, exchanges, orders, fills
  • Use your threading system (or be single-threaded)
  • Use your synchronization method

Who Should be Using Liquibook?

  • Exchanges that might want to distribute this as tool to their partners for their certification process

  • Traders starting to do exchange simulation -- they can also do backwards simulation of previous markets under certain conditions to review how strategies would have acted in the past
  • Traders doing predictive modeling
  • New smaller exchanges
  • High-frequency traders who want to simulate trading at market speeds or faster
WebSanity Top Secret