Threads are a fundamental part of the Java platform. As multicore processors become the norm, using concurrency effectively becomes essential for building high-performance applications. Java SE 5 and 6 are a huge step forward for the development of concurrent applications, with improvements to the Java Virtual Machine to support high-performance, highly scalable concurrent classes and a rich set of new concurrency building blocks. In Java Concurrency in Practice, the creators of these new facilities explain not only how they work and how to use them, but also the motivation and design patterns behind them.

However, developing, testing, and debugging multithreaded programs can still be very difficult; it is all too easy to create concurrent programs that appear to work, but fail when it matters most: in production, under heavy load. Java Concurrency in Practice arms readers with both the theoretical underpinnings and concrete techniques for building reliable, scalable, maintainable concurrent applications. Rather than simply offering an inventory of concurrency APIs and mechanisms, it provides design rules, patterns, and mental models that make it easier to build concurrent programs that are both correct and performant.

This book covers:

• Basic concepts of concurrency and thread safety
• Techniques for building and composing thread-safe classes
• Using the concurrency building blocks in java.util.concurrent
• Performance optimization dos and don'ts
• Testing concurrent programs
• Advanced topics such as atomic variables, nonblocking algorithms, and the Java Memory Model

Listings     xii
Preface     xvii

Chapter 1: Introduction     1

1.1  A (very) brief history of concurrency       1
1.2  Benefits of threads      3
1.3  Risks of threads       5
1.4  Threads are everywhere       9

2.1  What is thread safety?      17
2.2  Atomicity     19
2.3  Locking     23
2.4  Guarding state with locks      27
2.5  Liveness and performance       29

3.1  Visibility      33
3.2  Publication and escape       39
3.3  Thread confinement       42
3.4  Immutability       46
3.5  Safepublication       49

4.1  Designing a thread-safe class      55
4.2  Instance confinement      58
4.3  Delegating thread safety      62
4.4  Adding functionality to existing thread-safe classes       71
4.5  Documenting synchronization policies       74

5.1  Synchronized collections       79
5.2  Concurrent collections     84
5.3  Blocking queues and the producer-consumer pattern     87
5.4  Blocking and interruptible methods     92
5.5  Synchronizers     94
5.6  Building an efficient, scalable result cache      101

6.1  Executing tasks in threads      113
6.2  The Executor framework     117
6.3  Finding exploitable parallelism      123

7.1  Task cancellation      135
7.2  Stopping a thread-based service       150
7.3  Handling abnormal thread termination       161
7.4  JVM shutdown      164

8.1  Implicit couplings between tasks and execution policies     167
8.2  Sizing thread pools      170
8.3  Configuring ThreadPoolExecutor     171
8.4  Extending ThreadPoolExecutor     179
8.5  Parallelizing recursive algorithms     181

9.1  Why are GUIs single-threaded?      189
9.2  Short-running GUI tasks     192
9.3  Long-running GUI tasks     195
9.4  Shared data models     198
9.5  Other forms of single-threaded subsystems      202

10.1  Deadlock     205
10.2  Avoiding and diagnosing deadlocks     215
10.3  Other liveness hazards      218

11.1  Thinking about performance      221
11.2  Amdahl's law     225
11.3  Costs introduced by threads     229
11.4  Reducing lock contention      232
11.5  Example: Comparing Map performance     242
11.6  Reducing context switch overhead      243

12.1  Testing for correctness     248
12.2  Testing for performance      260
12.3  Avoiding performance testing pitfalls       266
12.4  Complementary testing approaches     270

13.1  Lock and ReentrantLock      277
13.2  Performance considerations      282
13.3  Fairness      283
13.4  Choosing between synchronized and ReentrantLock      285
13.5  Read-write locks     286

14.1  Managing state dependence      291
14.2  Using condition queues      298
14.3  Explicit condition objects     306
14.4  Anatomy of a synchronizer     308
14.5  AbstractQueuedSynchronizer      311
14.6  AQS in java.util.concurrent synchronizer classes      314

15.1  Disadvantages of locking     319
15.2  Hardware support for concurrency      321
15.3  Atomic variable classes       324
15.4  Nonblocking algorithms      329

16.1  What is a memory model, and why would I want one?       337
16.2  Publication     344
16.3  Initialization safety     349

A.1  Class annotations     353
A.2  Field andmethod annotations      353