Book Image

Developing High-Frequency Trading Systems

By : Sebastien Donadio, Sourav Ghosh, Romain Rossier
5 (1)
Book Image

Developing High-Frequency Trading Systems

5 (1)
By: Sebastien Donadio, Sourav Ghosh, Romain Rossier

Overview of this book

The world of trading markets is complex, but it can be made easier with technology. Sure, you know how to code, but where do you start? What programming language do you use? How do you solve the problem of latency? This book answers all these questions. It will help you navigate the world of algorithmic trading and show you how to build a high-frequency trading (HFT) system from complex technological components, supported by accurate data. Starting off with an introduction to HFT, exchanges, and the critical components of a trading system, this book quickly moves on to the nitty-gritty of optimizing hardware and your operating system for low-latency trading, such as bypassing the kernel, memory allocation, and the danger of context switching. Monitoring your system’s performance is vital, so you’ll also focus on logging and statistics. As you move beyond the traditional HFT programming languages, such as C++ and Java, you’ll learn how to use Python to achieve high levels of performance. And what book on trading is complete without diving into cryptocurrency? This guide delivers on that front as well, teaching how to perform high-frequency crypto trading with confidence. By the end of this trading book, you’ll be ready to take on the markets with HFT systems.
Table of Contents (16 chapters)
1
Part 1: Trading Strategies, Trading Systems, and Exchanges
5
Part 2: How to Architect a High-Frequency Trading System
10
Part 3: Implementation of a High-Frequency Trading System

Using kernel bypass

In this section, we will discuss using the kernel bypass technique to improve the performance of User Datagram Protocol (UDP) sockets to process inbound market data updates from the exchanges and Transmission Control Protocol (TCP) sockets to send outbound order flow/requests to the exchange. Fundamentally, kernel bypass looks to eliminate the expensive context switches and mode switches between kernel mode and user mode as well as duplicate copying of data from the Network Interface Card (NIC) to user space, each of which ends up reducing the latency quite a bit.

Network processing driven by system calls/interrupts in the non-kernel bypass design, threads, or processes that want to read incoming data on UDP or TCP socket block on the read call, as described in the Understanding context switches – interrupt handling section in the previous chapter. That leads the blocked thread or process being context switched out, and then it is woken up by the interrupt...