Synchronization

Synchronization

When two or more threads need access to a shared resource, they need some way to ensure that the resource will be used by only one thread at a time. The  process by which this is achieved is called synchronization. As you will see, Java provides unique, language-level support for it.

 

Key to synchronization is the concept of the monitor (also called a semaphore). A monitor is an object that is used as a mutually exclusive lock, or mutex. Only one thread can own a monitor at a given time. When a thread acquires a lock, it is said to have entered the monitor. All other threads attempting to enter the locked monitor will be suspended until the first thread exits the monitor. These other threads are said to be waiting for the monitor. A thread that owns a monitor can reenter the same monitor if it so desires.

 

If you have worked with synchronization when using other languages, such as C or C++, you know that it can be a bit tricky to use. This is because most languages do not, themselves, support synchronization. Instead, to synchronize threads, your programs need to utilize operating system primitives. Fortunately, because Java implements synchronization through language elements, most of the complexity associated with synchronization has been eliminated.

 

You can synchronize your code in either of two ways. Both involve the use of the synchronized keyword, and both are examined here.

 

Using Synchronized Methods

Synchronization is easy in Java, because all objects have their own implicit monitor associated with them. To enter an object's monitor, just call a method that has been modified with the synchronized keyword. While a thread is inside a synchronized method, all other threads that try to call it (or any other synchronized method) on the same instance have to wait. To exit the monitor and relinquish control of the object to the next waiting thread, the owner of the monitor simply returns from the synchronized method.

 

To understand the need for synchronization, let's begin with a simple example that does not use it—but should. The following program has three simple classes. The first one, Callme, has a single method named call( ). The call( ) method takes a String parameter called msg. This method tries to print the msg string inside of square brackets. The interesting thing to notice is that after call( ) prints the opening bracket and the msg string, it calls Thread.sleep(1000), which pauses the current thread for one second.

 

The constructor of the next class, Caller, takes a reference to an instance of the Callme class and a String, which are stored in target and msg, respectively. The constructor also creates a new thread that will call this object's run( ) method. The thread is started immediately. The run( ) method of Caller calls the call( ) method on the target instance of Callme, passing in the msg string. Finally, the Synch class starts by creating a single instance of Callme, and three instances of Caller, each with a unique message string. The same instance of Callme is passed to each Caller.

 

// This program is not synchronized.

class Callme {

void call(String msg) {

System.out.print("[" + msg);

try {

Thread.sleep(1000);

} catch(InterruptedException e) {

System.out.println("Interrupted");

}

System.out.println("]");

}

}

class Caller implements Runnable {

String msg;

Callme target;

Thread t;

public Caller(Callme targ, String s) {

target = targ;

msg = s;

t = new Thread(this);

t.start();

}

public void run() {

target.call(msg);

}

}

class Synch {

public static void main(String args[]) {

Callme target = new Callme();

Caller ob1 = new Caller(target, "Hello");

Caller ob2 = new Caller(target, "Synchronized");

Caller ob3 = new Caller(target, "World");

// wait for threads to end

try {

ob1.t.join();

ob2.t.join();

ob3.t.join();

} catch(InterruptedException e) {

System.out.println("Interrupted");

}

}

}

 

Here is the output produced by this program:

Hello[Synchronized[World]

]

]

 

As you can see, by calling sleep( ), the call( ) method allows execution to switch to another thread. This results in the mixed-up output of the three message strings. In this program, nothing exists to stop all three threads from calling the same method, on the same object, at the same time. This is known as a race condition, because the three threads are racing each other to complete the method. This example used sleep( ) to make the effects repeatable and obvious. In most situations, a race condition is more subtle and less predictable, because you can't be sure when the context switch will occur. This can cause a program to run right one time and wrong the next.

 

To fix the preceding program, you must serialize access to call( ). That is, you must restrict its access to only one thread at a time. To do this, you simply need to precede call( )'s definition with the keyword synchronized, as shown here:

 

class Callme {

synchronized void call(String msg) {

 

This prevents other threads from entering call( ) while another thread is using it.

After synchronized has been added to call( ), the output of the program is as follows:

 

[Hello]

[Synchronized]

[World]

 

Any time that you have a method, or group of methods, that manipulates the internal state of an object in a multithreaded situation, you should use the synchronized keyword to guard the state from race conditions. Remember, once a thread enters any synchronized method on an instance, no other thread can enter any other synchronized method on the same instance. However, nonsynchronized methods on that instance will continue to be callable.