Limit Order Book & Matching Engine

This repository contains the C++ implementation of a high-performance Limit Order Book and its Matching Engine.

Overview

The Limit Order Book is designed using a combination of data structures to optimize order insertion, deletion, and matching:

Architecture

• Data Structures: Orders are stored in a doubly linked-list structure, where each list represents a price level. Price levels are managed using Red-Black Trees implemented via std::set in C++, ensuring efficient insertion and deletion operations.

• Order Format: Orders are added in the format ID, AgentID, Price, Volume, and deleted by their id ID.

Performance

To assess performance, the system was tested with 1,000,000 operations (place or delete orders), generated using a Python script. These operations were processed through a CSV parser and fed into the C++ program.

• High-Performance: The system achieved a peak performance of 600,000 operations per second. Operations include order insertions, trades, and order deletions.

Build Instructions

Prerequisites

• CMake 3.28 or higher
• C++20 compatible compiler
• Google Test (for running tests)

Building the Project

mkdir build
cd build
cmake ..
make

Running Tests

cd build
make
./tests/Google_Tests_run

Usage

The program takes two command-line arguments: an input CSV file and an output CSV file.

./OrderBook_run <input_file.csv> <output_file.csv>

Input Format

The input CSV file should contain operations in the following format:

Operation,OrderID,AgentID,Side,Price,Volume
PLACE,1,0,0,100,50
PLACE,2,0,1,101,30
DELETE,1,,,

• PLACE: Add a new order (requires OrderID, AgentID, Side, Price, Volume)
  • Side: 0 = BUY, 1 = SELL
• DELETE: Remove an order (requires only OrderID)

Output Format

The output CSV will contain the final order book state:

Price Limit,Side,Volume
100,BUY,50
101,SELL,30

Demo

For a demonstration, refer to the Python script used to generate sample orders. You can run it to create test data:

python demo/generate_orders.py
./OrderBook_run demo/sample_operations.csv demo/final_order_book.csv

References

1. How to Build a Fast Limit Order Book - Insights into designing performant limit order books.

---

This project demonstrates a robust implementation of a Limit Order Book and Matching Engine in C++, optimized for high-frequency trading environments.