Concurrency with Modern C++
Concurrency with Modern C++
What every professional C++ programmer should know about concurrency.
About the Book
- C++11 and C++14 have the basic building blocks for creating concurrent or parallel programs.
- With C++17, we got the parallel algorithms of the Standard Template Library (STL). That means, most of the algorithms of the STL can be executed sequential, parallel, or vectorized.
- The concurrency story in C++ goes on. With C++20, we got coroutines, atomic smart pointers, semaphores, and latches and barriers.
- With C++23, we can hope for executors, extended futures, transactional memory, and more.
This book explains the details of concurrency in modern C++ and gives you, also, nearly 200 running code examples. Therefore, you can combine the theory with the practices and get the most of it.
Because this book is about concurrency, I present many pitfalls and show you how to overcome them.
The book is 100 % finished, but I will update it from time to time. The next update is probably about C++23. Furthermore, I will start the chapter on lock-free data structures.
About the Author
Packages
The Book
PDF
EPUB
WEB
English
Concurreny with Modern C++ Team Edition: Five Copies
Get five copies to the price of three. This package includes all code examples.
PDF
EPUB
WEB
English
Bundles that include this book
Reader Testimonials
Bart Vandewoestyne
Senior Development Engineer Software at Esterline
'Concurrency with Modern C++' is your practical guide to getting familiar with concurrent programming in Modern C++. Starting with the C++ Memory Model and using many ready-to-run code examples, the book covers a good deal of what you need to improve your C++ multithreading skills. Next to the enlightening case studies that will bring you up to speed, the overview of upcoming concurrency features might even wet your appetite for more!
Ian Reeve
Senior Storage Software Engineer for Dell Inc.
Rainer Grimm's Concurrency with Modern C++ is a well written book covering the theory and practice for working with concurrency per the existing C++ standards, as well as addressing the potential changes for the upcoming C++ 20 standard. He provides a conversational discussion of the applications and best practices for concurrency along with example code to reinforce the details of each topic. An informative and worthwhile read!
Robert Badea
Technical Team Leader
Concurrency with Modern C++ is the easiest way to become an expert in the multithreading environment. This book contains both simple and advanced topics, and it has everything a developer needs, in order to become an expert in this field: Lots of content, a big number of running code examples, along with great explanation, and a whole chapter for pitfalls. I enjoyed reading it, and I highly recommend it for everyone working with C++.
Zeshuang Mi
Postgraduate
Concurrency with Modern C++ makes the multithreading's concept so clear. At the beginning of this book, it makes me know very clearly about memory model. The memory model is very important to the multithreading. The book makes me have confidence when i write my concurrent program, which makes me know what should i do or what i shouldn't do. The author elaborates the multithreading pattern which makes so helpful to me. I love it so much.
Table of Contents
-
- Reader Testimonials
-
Introduction
-
Conventions
- Special Fonts
- Special Symbols
- Special Boxes
-
Source Code
- Run the Programs
- How should you read the book?
-
Personal Notes
- Acknowledgment
- About Me
- My Special Circumstances
-
Conventions
-
A Quick Overview
-
1. Concurrency with Modern C++
-
1.1 C++11 and C++14: The Foundation
- 1.1.1 Memory Model
- 1.1.2 Multithreading
-
1.2 C++17: Parallel Algorithms of the Standard Template Library
- 1.2.1 Execution Policy
- 1.2.2 New Algorithms
- 1.3 Coroutines
-
1.4 Case Studies
- 1.4.1 Calculating the Sum of a Vector
- 1.4.2 The Dining Philosophers Problem by Andre Adrian
- 1.4.3 Thread-Safe Initialization of a Singleton
- 1.4.4 Ongoing Optimization with CppMem
- 1.4.5 Fast Synchronization of Threads
- 1.5 Variations of Futures
- 1.6 Modification and Generalization of a Generator
- 1.7 Various Job Workflows
-
1.8 The Future: C++23
- 1.8.1 Executors
- 1.8.2 Extended futures
- 1.8.3 Transactional Memory
- 1.8.4 Task Blocks
- 1.8.5 Data-Parallel Vector Library
-
1.9 Patterns and Best Practices
- 1.9.1 Synchronization
- 1.9.2 Concurrent Architecture
- 1.9.3 Best Practices
- 1.10 Data Structures
- 1.11 Challenges
- 1.12 Time Library
- 1.13 CppMem
- 1.14 Glossary
-
1.1 C++11 and C++14: The Foundation
-
1. Concurrency with Modern C++
-
The Details
-
2. Memory Model
-
2.1 Basics of the Memory Model
- 2.1.1 What is a memory location?
- 2.1.2 What happens if two threads access the same memory location?
-
2.2 The Contract
- 2.2.1 The Foundation
- 2.2.2 The Challenges
-
2.3 Atomics
- 2.3.1 Strong versus Weak Memory Model
- 2.3.2 The Atomic Flag
-
2.3.3
std::atomic - 2.3.4 All Atomic Operations
- 2.3.5 Free Atomic Functions
-
2.3.6
std::atomic_ref(C++20)
-
2.4 The Synchronization and Ordering Constraints
- 2.4.1 The Six Variants of Memory Orderings in C++
- 2.4.2 Sequential Consistency
- 2.4.3 Acquire-Release Semantic
-
2.4.4
std::memory_order_consume - 2.4.5 Relaxed Semantic
-
2.5 Fences
-
2.5.1
std::atomic_thread_fence -
2.5.2
std::atomic_signal_fence
-
2.5.1
-
2.1 Basics of the Memory Model
-
3. Multithreading
-
3.1 The Basic Thread
std::thread- 3.1.1 Thread Creation
- 3.1.2 Thread Lifetime
- 3.1.3 Thread Arguments
- 3.1.4 Member Functions
-
3.2 The Improved Thread
std::jthread(C++20)- 3.2.1 Automatically Joining
-
3.2.2 Cooperative Interruption of a
std::jthread
-
3.3 Shared Data
- 3.3.1 Mutexes
- 3.3.2 Locks
-
3.3.3
std::lock - 3.3.4 Thread-safe Initialization
- 3.4 Thread-Local Data
-
3.5 Condition Variables
- 3.5.1 The Predicate
- 3.5.2 Lost Wakeup and Spurious Wakeup
- 3.5.3 The Wait Workflow
-
3.6 Cooperative Interruption (C++20)
-
3.6.1
std::stop_source -
3.6.2
std::stop_token -
3.6.3
std::stop_callback - 3.6.4 A General Mechanism to Send Signals
-
3.6.5 Additional Functionality of
std::jthread -
3.6.6 New
waitOverloads for thecondition_variable_any
-
3.6.1
- 3.7 Semaphores (C++20)
-
3.8 Latches and Barriers (C++20)
-
3.8.1
std::latch -
3.8.2
std::barrier
-
3.8.1
-
3.9 Tasks
- 3.9.1 Tasks versus Threads
-
3.9.2
std::async -
3.9.3
std::packaged_task -
3.9.4
std::promiseandstd::future -
3.9.5
std::shared_future - 3.9.6 Exceptions
- 3.9.7 Notifications
- 3.10 Synchronized Outputstreams (C++20)
-
3.1 The Basic Thread
-
4. Parallel Algorithms of the Standard Template Library
-
4.1 Execution Policies
- 4.1.1 Parallel and Vectorized Execution
- 4.1.2 Exceptions
- 4.1.3 Hazards of Data Races and Deadlocks
- 4.2 Algorithms
-
4.3 The New Algorithms
- 4.3.1 More overloads
- 4.3.2 The functional Heritage
-
4.4 Compiler Support
- 4.4.1 Microsoft Visual Compiler
- 4.4.2 GCC Compiler
- 4.4.3 Further Implementations of the Parallel STL
-
4.5 Performance
- 4.5.1 Microsoft Visual Compiler
- 4.5.2 GCC Compiler
-
4.1 Execution Policies
-
5. Coroutines (C++20)
- 5.1 A Generator Function
-
5.2 Characteristics
- 5.2.1 Typical Use Cases
- 5.2.2 Underlying Concepts
- 5.2.3 Design Goals
- 5.2.4 Becoming a Coroutine
-
5.3 The Framework
- 5.3.1 Promise Object
- 5.3.2 Coroutine Handle
- 5.3.3 Coroutine Frame
-
5.4 Awaitables and Awaiters
- 5.4.1 Awaitables
- 5.4.2 The Concept Awaiter
-
5.4.3
std::suspend_alwaysandstd::suspend_never -
5.4.4
initial_suspend -
5.4.5
final_suspend - 5.4.6 Awaiter
-
5.5 The Workflows
- 5.5.1 The Promise Workflow
- 5.5.2 The Awaiter Workflow
-
5.6
co_return- 5.6.1 A Future
-
5.7
co_yield- 5.7.1 An Infinite Data Stream
-
5.8
co_await- 5.8.1 Starting a Job on Request
- 5.8.2 Thread Synchronization
-
6. Case Studies
-
6.1 Calculating the Sum of a Vector
- 6.1.1 Single Threaded addition of a Vector
- 6.1.2 Multithreaded Summation with a Shared Variable
- 6.1.3 Thread-Local Summation
- 6.1.4 Summation of a Vector: The Conclusion
-
6.2 The Dining Philosophers Problem by Andre Adrian
- 6.2.1 Multiple Resource Use
- 6.2.2 Multiple Resource Use with Logging
- 6.2.3 Erroneous Busy Waiting without Resource Hierarchy
- 6.2.4 Erroneous Busy Waiting with Resource Hierarchy
- 6.2.5 Still Erroneous Busy Waiting with Resource Hierarchy
- 6.2.6 Correct Busy Waiting with Resource Hierarchy
- 6.2.7 Good low CPU load Busy Waiting with Resource Hierarchy
-
6.2.8
std::mutexwith Resource Hierarchy -
6.2.9
std::lock_guardwith Resource Hierarchy -
6.2.10
std::lock_guardand Synchronized Output with Resource Hierarchy -
6.2.11
std::lock_guardand Synchronized Output with Resource Hierarchy and a count -
6.2.12 A
std::unique_lockusing deferred locking -
6.2.13 A
std::scoped_lockwith Resource Hierarchy - 6.2.14 The Original Dining Philosophers Problem using Semaphores
- 6.2.15 A C++20 Compatible Semaphore
-
6.3 Thread-Safe Initialization of a Singleton
- 6.3.1 Double-Checked Locking Pattern
- 6.3.2 Performance Measurement
- 6.3.3 Thread-Safe Meyers Singleton
-
6.3.4
std::lock_guard -
6.3.5
std::call_oncewithstd::once_flag - 6.3.6 Atomics
- 6.3.7 Performance Numbers of the various Thread-Safe Singleton Implementations
-
6.4 Ongoing Optimization with CppMem
- 6.4.1 CppMem: Non-Atomic Variables
- 6.4.2 CppMem: Locks
- 6.4.3 CppMem: Atomics with Sequential Consistency
- 6.4.4 CppMem: Atomics with Acquire-Release Semantic
- 6.4.5 CppMem: Atomics with Non-atomics
- 6.4.6 CppMem: Atomics with Relaxed Semantic
- 6.4.7 Conclusion
-
6.5 Fast Synchronization of Threads
- 6.5.1 Condition Variables
-
6.5.2
std::atomic_flag -
6.5.3
std::atomic<bool> - 6.5.4 Semaphores
- 6.5.5 All Numbers
-
6.6 Variations of Futures
- 6.6.1 A Lazy Future
- 6.6.2 Execution on Another Thread
-
6.7 Modification and Generalization of a Generator
- 6.7.1 Modifications
- 6.7.2 Generalization
-
6.8 Various Job Workflows
- 6.8.1 The Transparent Awaiter Workflow
- 6.8.2 Automatically Resuming the Awaiter
- 6.8.3 Automatically Resuming the Awaiter on a Separate Thread
-
6.1 Calculating the Sum of a Vector
-
7. The Future: C++23
-
7.1 Executors
- 7.1.1 A long Way
- 7.1.2 What is an Executor?
- 7.1.3 First Examples
- 7.1.4 Goals of an Executor Concept
- 7.1.5 Terminology
- 7.1.6 Execution Functions
- 7.1.7 A Prototype Implementation
-
7.2 Extended Futures
- 7.2.1 Concurrency TS v1
- 7.2.2 Unified Futures
-
7.3 Transactional Memory
- 7.3.1 ACI(D)
- 7.3.2 Synchronized and Atomic Blocks
-
7.3.3
transaction_safeversustransaction_unsafeCode
-
7.4 Task Blocks
- 7.4.1 Fork and Join
-
7.4.2
define_task_blockversusdefine_task_block_restore_thread - 7.4.3 The Interface
- 7.4.4 The Scheduler
-
7.5 Data-Parallel Vector Library
- 7.5.1 Data-Parallel Vectors
- 7.5.2 The Interface of the Data-Parallel Vectors
-
7.1 Executors
-
2. Memory Model
-
Patterns
-
8. Patterns and Best Practices
- 8.1 History
- 8.2 Invaluable Value
- 8.3 Pattern versus Best Practices
- 8.4 Anti-Pattern
-
9. Synchronization Patterns
-
9.1 Dealing with Sharing
- 9.1.1 Copied Value
- 9.1.2 Thread-Specific Storage
- 9.1.3 Future
-
9.2 Dealing with Mutation
- 9.2.1 Scoped Locking
- 9.2.2 Strategized Locking
- 9.2.3 Thread-Safe Interface
- 9.2.4 Guarded Suspension
-
9.1 Dealing with Sharing
-
10. Concurrent Architecture
-
10.1 Active Object
- 10.1.1 Challenges
- 10.1.2 Solution
- 10.1.3 Components
- 10.1.4 Dynamic Behavior
- 10.1.5 Advantages and Disadvantages
- 10.1.6 Implementation
-
10.2 Monitor Object
- 10.2.1 Challenges
- 10.2.2 Solution
- 10.2.3 Components
- 10.2.4 Dynamic Behavior
- 10.2.5 Advantages and Disadvantages
-
10.3 Half-Sync/Half-Async
- 10.3.1 Challenges
- 10.3.2 Solution
- 10.3.3 Components
- 10.3.4 Dynamic Behavior
- 10.3.5 Advantages and Disadvantages
- 10.3.6 Example
-
10.4 Reactor
- 10.4.1 Challenges
- 10.4.2 Solution
- 10.4.3 Components
- 10.4.4 Dynamic Behavior
- 10.4.5 Advantages and Disadvantages
- 10.4.6 Example
-
10.5 Proactor
- 10.5.1 Challenges
- 10.5.2 Solution
- 10.5.3 Components
- 10.5.4 Advantages and Disadvantages
- 10.5.5 Example
- 10.6 Further Information
-
10.1 Active Object
-
11. Best Practices
-
11.1 General
- 11.1.1 Code Reviews
- 11.1.2 Minimize Sharing of Mutable Data
- 11.1.3 Minimize Waiting
- 11.1.4 Prefer Immutable Data
- 11.1.5 Use pure functions
- 11.1.6 Look for the Right Abstraction
- 11.1.7 Use Static Code Analysis Tools
- 11.1.8 Use Dynamic Enforcement Tools
-
11.2 Multithreading
- 11.2.1 Threads
- 11.2.2 Data Sharing
- 11.2.3 Condition Variables
- 11.2.4 Promises and Futures
-
11.3 Memory Model
- 11.3.1 Don’t use volatile for synchronization
- 11.3.2 Don’t program Lock Free
- 11.3.3 If you program Lock-Free, use well-established patterns
- 11.3.4 Don’t build your abstraction, use guarantees of the language
- 11.3.5 Don’t reinvent the wheel
-
11.1 General
-
8. Patterns and Best Practices
-
Data Structures
-
12. General Considerations
- 12.1 Concurrent Stack
- 12.2 Locking Strategy
- 12.3 Granularity of the Interface
-
12.4 Typical Usage Pattern
- 12.4.1 Linux (GCC)
- 12.4.2 Windows (cl.exe)
- 12.5 Avoidance of Loopholes
-
12.6 Contention
- 12.6.1 Single-Threaded Summation without Synchronization
- 12.6.2 Single-Threaded Summation with Synchronization (lock)
- 12.6.3 Single-Threaded Summation with Synchronization (atomic)
- 12.6.4 The Comparison
- 12.7 Scalability
- 12.8 Invariants
- 12.9 Exceptions
-
13. Lock-Based Data Structures
-
- 13.0.1 A Stack
-
13.1 Concurrent Queue
- 13.1.1 A Queue
- 13.1.2 Coarse-Grained Locking
- 13.1.3 Fine-Grained Locking
-
-
14. Lock-Free Data Structures
-
14.1 General Considerations
- 14.1.1 The Next Evolutionary Step
- 14.1.2 Sequential Consistency
-
14.2 Concurrent Stack
- 14.2.1 A Simplified Implementation
- 14.2.2 A Complete Implementation
- 14.3 Concurrent Queue
-
14.1 General Considerations
-
12. General Considerations
-
Further Information
-
15. Challenges
- 15.1 ABA Problem
- 15.2 Blocking Issues
- 15.3 Breaking of Program Invariants
- 15.4 Data Races
- 15.5 Deadlocks
- 15.6 False Sharing
- 15.7 Lifetime Issues of Variables
- 15.8 Moving Threads
- 15.9 Race Conditions
-
16. The Time Library
- 16.1 The Interplay of Time Point, Time Duration, and Clock
-
16.2 Time Point
- 16.2.1 From Time Point to Calendar Time
- 16.2.2 Cross the valid Time Range
-
16.3 Time Duration
- 16.3.1 Calculations
-
16.4 Clocks
- 16.4.1 Accuracy and Steadiness
- 16.4.2 Epoch
- 16.5 Sleep and Wait
-
17. CppMem - An Overview
-
17.1 The simplified Overview
- 17.1.1 1. Model
- 17.1.2 2. Program
- 17.1.3 3. Display Relations
- 17.1.4 4. Display Layout
- 17.1.5 5. Model Predicates
- 17.1.6 The Examples
-
17.1 The simplified Overview
-
18. Glossary
-
18.1
adress_free - 18.2 ACID
- 18.3 CAS
- 18.4 Callable Unit
- 18.5 Complexity
- 18.6 Concepts
- 18.7 Concurrency
- 18.8 Critical Section
- 18.9 Deadlock
- 18.10 Eager Evaluation
- 18.11 Executor
- 18.12 Function Objects
- 18.13 Lambda Functions
- 18.14 Lazy evaluation
- 18.15 Lock-free
- 18.16 Lock-based
- 18.17 Lost Wakeup
- 18.18 Math Laws
- 18.19 Memory Location
- 18.20 Memory Model
- 18.21 Modification Order
- 18.22 Monad
- 18.23 Non-blocking
- 18.24 obstruction-free
- 18.25 Parallelism
- 18.26 Predicate
- 18.27 Pattern
- 18.28 RAII
- 18.29 Release Sequence
- 18.30 Sequential Consistency
- 18.31 Sequence Point
- 18.32 Spurious Wakeup
- 18.33 Thread
- 18.34 Total order
- 18.35 TriviallyCopyable
- 18.36 Undefined Behavior
- 18.37 volatile
- 18.38 wait-free
-
18.1
- Index
-
15. Challenges
The Leanpub 60-day 100% Happiness Guarantee
Within 60 days of purchase you can get a 100% refund on any Leanpub purchase, in two clicks.
See full terms
Do Well. Do Good.
Authors have earned$11,955,464writing, publishing and selling on Leanpub, earning 80% royalties while saving up to 25 million pounds of CO2 and up to 46,000 trees.
Learn more about writing on Leanpub
Free Updates. DRM Free.
If you buy a Leanpub book, you get free updates for as long as the author updates the book! Many authors use Leanpub to publish their books in-progress, while they are writing them. All readers get free updates, regardless of when they bought the book or how much they paid (including free).
Most Leanpub books are available in PDF (for computers) and EPUB (for phones, tablets and Kindle). The formats that a book includes are shown at the top right corner of this page.
Finally, Leanpub books don't have any DRM copy-protection nonsense, so you can easily read them on any supported device.
Learn more about Leanpub's ebook formats and where to read them
Top Books
- #1
OpenIntro Statistics
David Diez, Christopher Barr, Mine Cetinkaya-Rundel, and OpenIntroA complete foundation for Statistics, also serving as a foundation for Data Science.
Leanpub revenue supports OpenIntro (US-based nonprofit) so we can provide free desk copies to teachers interested in using OpenIntro Statistics in the classroom and expand the project to support free textbooks in other subjects.
More resources: openintro.org.
- #2
Deep Learning with PyTorch Step-by-Step
Daniel Voigt GodoyIn 2019, I published a PyTorch tutorial on Towards Data Science and I was amazed by the reaction from the readers! Their feedback motivated me to write this book to help beginners start their journey into Deep Learning and PyTorch. I hope you enjoy reading this book as much as I enjoy writing it.
- #3
Linear Algebra
Elira Curri and Tanush ShaskaAnswers to selected problems included.
- #4
Modern IT Automation with PowerShell
The DevOps Collective, Inc. and Michael ZanattaA PowerShell Textbook written by the community for the community!
- #5
Ansible for DevOps
Jeff GeerlingAnsible is a simple, but powerful, server and configuration management tool. Learn to use Ansible effectively, whether you manage one server—or thousands.
- #6
R Programming for Data Science
Roger D. PengThis book brings the fundamentals of R programming to you, using the same material developed as part of the industry-leading Johns Hopkins Data Science Specialization. The skills taught in this book will lay the foundation for you to begin your journey learning data science. Printed copies of this book are available through Lulu.
- #7
Code Your Own Synth Plug-Ins With C++ and JUCE
Matthijs HollemansLearn the fundamentals of audio programming by building a fully-featured software synthesizer plug-in, with every step explained along the way. The plug-in can be used in all the popular DAWs and is made using the industry standard tools for audio development: JUCE framework and the C++ programming language. Not too much math, lots of explanations!
- #8
The Rails 7 Way
Obie Fernandez, Lucas Dohmen, and Tom Henrik AadlandThe Rails™ 7 Way is the comprehensive, authoritative reference guide for professionals delivering production-quality code using modern Ruby on Rails. It illuminates the entire Rails 7 API, its most powerful idioms, design approaches, and libraries. Building on the previous editions, this edition has been heavily refactored and updated.
- #9
Software Architecture for Developers
Simon BrownA developer-friendly, practical and pragmatic guide to lightweight software architecture, technical leadership and the balance with agility.
- #10
node-opcua by example
Etienne RossignonGet the best out of node-opcua through a set of documented examples by the author himself that will allow you to create stunning OPCUA Servers or Clients.
Top Bundles
- #1
Software Architecture
2 Books
"Software Architecture for Developers" is a practical and pragmatic guide to modern, lightweight software architecture, specifically aimed at developers. You'll learn:The essence of software architecture.Why the software architecture role should include coding, coaching and collaboration.The things that you really need to think about before... - #2
CCIE Service Provider Ultimate Study Bundle
2 Books
Piotr Jablonski, Lukasz Bromirski, and Nick Russo have joined forces to deliver the only CCIE Service Provider training resource you'll ever need. This bundle contains a detailed and challenging collection of workbook labs, plus an extensively detailed technical reference guide. All of us have earned the CCIE Service Provider certification... - #3
Pattern-Oriented Memory Forensics and Malware Detection
2 Books
This training bundle for security engineers and researchers, malware and memory forensics analysts includes two accelerated training courses for Windows memory dump analysis using WinDbg. It is also useful for technical support and escalation engineers who analyze memory dumps from complex software environments and need to check for possible... - #4
Static Analysis and Automated Refactoring
2 Books
As PHP developers we are living in the "Age of Static Analysis". We can use a tool like PHPStan to learn about potential bugs before we ship our code to production, and we can enforce our team's programming standards using custom PHPStan rules. Recipes for Decoupling by Matthias Noback teaches you in great detail how to do this, while also... - #5
Practical FP in Scala + Functional event-driven architecture
2 Books
Practical FP in Scala (A hands-on approach) & Functional event-driven architecture, aka FEDA, (Powered by Scala 3), together as a bundle! The content of PFP in Scala is a requirement to understand FEDA so why not take advantage of this bundle!? - #6
Modern C++ Collection
3 Books
Get All about Modern C++C++ Standard Library, including C++20Concurrency with Modern C++, including C++20C++20Each book has about 200 complete code examples. Updates are included. When I update one of the books, you immediately get the updated bundle. You can expect significant updates to each new C++ standard (C++23, C++26, .. ) and also... - #7
All the Books of The Medical Futurist
6 Books
We put together the most popular books from The Medical Futurist to provide a clear picture about the major trends shaping the future of medicine and healthcare. Digital health technologies, artificial intelligence, the future of 20 medical specialties, big pharma, data privacy, digital health investments and how technology giants such as Amazon... - #9
Statistics - All Lee Baker's Books
10 Books
This bundle includes all of the statistics books that Lee Baker has published at LeanPub 13 Statistics Books*1 PlaceNo Hassle! All in both epub and mobi formats - perfect for all devices and screen sizes. --------------------------------------------------------- *The book listed as Getting Started With Statistics is a bundle of 4 books...
