Developing High Frequency Trading Systems: Learn how to implement high-frequency trading from scratch with C++ or Java basics

Cover
Title Page
Copyright and Credits
Contributors
Table of Contents
Preface
Part 1: Trading Strategies, Trading Systems, and Exchanges
Chapter 1: Fundamentals of a High-Frequency Trading System
	History of HFT
		The post-1930s era
		The modern era
	Why have HFT?
	What makes HFT so different from regular trading?
		Effect of dark pools
	Who trades HFT?
	What do I need to start an HFT?
		What are HFT strategies?
		Asset classes
		Liquidity
		Tick-by-tick data and data distribution
		Liquidity rebates
		Matching engine
		Market making
		Scalping
		Statistical arbitrage
		Latency arbitrage
		Impact of news
		Momentum ignition
		Rebate strategies
		Pinging
		Illegal activities
	Summary
Chapter 2: The Critical Components of a Trading System
	Understanding the trading system
		Trading system architecture
		Gateways connecting to trading exchanges
	Making a trading system trade with exchanges
		Examining the API for communication
	Order book management
		Order book considerations
	Strategy making decisions on when to trade
		The OMS
		Critical components
		Non-critical components
		Command and control
		Services
	Summary
Chapter 3: Understanding the Trading Exchange Dynamics
	Architecting a trading exchange for handling orders at a large scale
		History of trading exchanges
		Understanding features of an exchange
		Exchange architecture
	General order book and matching engine
		Best price scenario
		Partial fill scenario
		No match scenario
		Multiple orders with the same price
	Summary
Part 2: How to Architect a High-Frequency Trading System
Chapter 4: HFT System Foundations – From Hardware to OS
	Understanding HFT computers
		CPUs, from multi-processor to multi-core
		Main memory or RAM
		Shared memory
		I/O devices
	Using the OS for HFT systems
		User space and kernel space
		Process scheduling and CPU resource management
		Memory management
		Paged memory and page tables
		System calls
		Threading
		Interruption management
	The role of compilers
		Executable file formats
		Static versus dynamic linking
	Summary
Chapter 5: Networking in Motion
	Understanding networking in HFT systems
		Learning about network conceptual models
	Network communications between systems in HFT
		Comprehending how switches work
	Important protocol concepts
	Using Ethernet for HFT communication
		Using IPv4 as a network layer
		UDP and TCP for the transport layer
	Designing financial protocols for HFT exchanges
		FIX protocol
	Interior networks versus exterior networks
	Understanding the packet life cycle
		Comprehending the packet life in the send/receive (TX/RX) path
		Software layer receiving the packet
	Monitoring the network
		Packet capture and analysis
	Valuing time distribution
		Time-synchronization services
	Summary
Chapter 6: HFT Optimization – Architecture and Operating System
	Performance mental model
	Understanding context switches 
		Types of context switches
		Why are context switches good
		Steps and operations involved in a context switch operation
		Why are context switches bad for HFT?
		Techniques to avoid or minimize context switches
	Building lock-free data structures 
		When/why are locks needed (non-HFT applications)
		Types of synchronization mechanisms
		Problems and inefficiencies with using locks
	Pre-fetching and pre-allocating memory
		Memory hierarchy
		Pre-fetching based alternatives to boost performance
		Dynamic memory allocation
		Pre-allocation-based alternatives to dynamic memory allocation
	Summary
Chapter 7: HFT Optimization – Logging, Performance, and Networking
	Comparing kernel space and user space 
		What is kernel and user space?
		Investigating performance – kernel versus user space
	Using kernel bypass
		Understanding why kernel bypass is the alternative
		Presenting kernel bypass latencies
	Learning about memory-mapped files
	Using cable fiber, hollow fiber, and microwave technologies 
		Evolution from cable fiber to hollow fiber to microwave
		How hollow fiber works
		How microwave works
	Diving into logging and statistics
		The need for logging in HFT
		The need for online/live statistics computation in HFT
	Measuring performance
		Motivation for measuring performance
		Linux tools for measuring performance
		Custom techniques for measuring performance
	Summary
Part 3: Implementation of a High-Frequency Trading System
Chapter 8: C++ – The Quest for Microsecond Latency
	C++ 14/17 memory model 
		What is a memory model?
		The need for a memory model
		The C++ 11 memory model and its rules
		C++ memory model principles
	Removing runtime decisions 
		Motivation for removing runtime decisions
		Virtual functions
		Performance penalties
	Dynamic memory allocation 
		Runtime performance penalty
		Using constexpr efficiently {ICON2}
		Exceptions impeding performance {ICON2}
	Templates reducing the runtime 
		What are templates?
		Template specialization {ICON2}
		Why use templates?
		Disadvantages of templates
		Performance of templates
		Standard Template Library (STL)
	Static analysis
		What is C++ static analysis?
		The need for static analysis
		Types of static analysis
		Steps in static analysis
		Benefits and drawbacks of static analysis
	Use case-Building an FX high-frequency trading system
	Summary
Chapter 9: Java and JVM for Low-Latency Systems
	Introducing the basics of Java
	Reducing the impact of the GC 
		What to do to keep GC events low and fast
	Warming up the JVM 
		Tiered compilation in JVM
		Optimizing the JVM for better startup performance
	Measuring the performance of a Java software
		Why are Java microbenchmarks difficult to create?
		Real-time performance measures
	Java threading 
		Using a thread pool/queue with threads
	High-performance task queue 
		Queues
		Circular buffer
		LMAX disruptor
	Logging and DB access
		External or internal thread?
	Summary
Chapter 10: Python – Interpreted but Open to High Performance
	Introducing Python
		Making use of Python for analytics
		Why is Python slow?
		How do we use libraries in Python?
	Python and C++ for HFT
		Using C++ in Python
		Using Python with C++
		Boost.Python library
		Using ctypes/CFFI to accelerate Python code
		SWIG
	Improving the speed of Python code in HFT
	Summary
Chapter 11: High-Frequency FPGA and Crypto
	Reducing latencies with FPGA 
		Evolution of the fierce competition of speed in HFT
		Introduction to FPGA
		Diving into FPGA trading systems
		Advantages of FPGA trading systems
		Disadvantages of FPGA trading systems
		Final words on FPGAs
	Exploring HFT with cryptocurrencies
		What is crypto?
		How do crypto transactions work?
		What is a blockchain?
		What is cryptocurrency mining?
		Similarities between traditional asset trading and cryptocurrency trading
		Main differences between traditional asset trading and cryptocurrency trading
		Trading with cryptocurrency exchange
		HFT strategies in crypto
		Building a high-frequency system for crypto trading
		How to build a trading system in the cloud
	Summary
Index
Other Books You May Enjoy