Semaphores in Java
2. Semaphore
We’ll start with java.util.concurrent.Semaphore. We can use semaphores to limit the number of concurrent threads accessing a specific resource.
In the following example, we will implement a simple login queue to limit number users in the system:
class LoginQueueUsingSemaphore {
private Semaphore semaphore;
public LoginQueueUsingSemaphore(int slotLimit) {
semaphore = new Semaphore(slotLimit);
}
boolean tryLogin() {
return semaphore.tryAcquire();
}
void logout() {
semaphore.release();
}
int availableSlots() {
return semaphore.availablePermits();
}
}
Notice how we used the following methods:
-
tryAcquire() – return true if a permit is available immediately and acquire it otherwise return false, but acquire() acquires a permit and blocking until one is available
-
release() – release a permit
-
availablePermits() – return number of current permits available
To test our login queue, we will first try to reach the limit and check if the next login attempt will be blocked:
@Test
public void givenLoginQueue_whenReachLimit_thenBlocked() {
int slots = 10;
ExecutorService executorService = Executors.newFixedThreadPool(slots);
LoginQueueUsingSemaphore loginQueue = new LoginQueueUsingSemaphore(slots);
IntStream.range(0, slots)
.forEach(user -> executorService.execute(loginQueue::tryLogin));
executorService.shutdown();
assertEquals(0, loginQueue.availableSlots());
assertFalse(loginQueue.tryLogin());
}
Next, we will see if any slots are available after a logout:
@Test
public void givenLoginQueue_whenLogout_thenSlotsAvailable() {
int slots = 10;
ExecutorService executorService = Executors.newFixedThreadPool(slots);
LoginQueueUsingSemaphore loginQueue = new LoginQueueUsingSemaphore(slots);
IntStream.range(0, slots)
.forEach(user -> executorService.execute(loginQueue::tryLogin));
executorService.shutdown();
assertEquals(0, loginQueue.availableSlots());
loginQueue.logout();
assertTrue(loginQueue.availableSlots() > 0);
assertTrue(loginQueue.tryLogin());
}
3. Timed Semaphore
Next, we will discuss Apache Commons TimedSemaphore. TimedSemaphore allows a number of permits as a simple Semaphore but in a given period of time, after this period the time reset and all permits are released.
We can use TimedSemaphore to build a simple delay queue as follows:
class DelayQueueUsingTimedSemaphore {
private TimedSemaphore semaphore;
DelayQueueUsingTimedSemaphore(long period, int slotLimit) {
semaphore = new TimedSemaphore(period, TimeUnit.SECONDS, slotLimit);
}
boolean tryAdd() {
return semaphore.tryAcquire();
}
int availableSlots() {
return semaphore.getAvailablePermits();
}
}
When we use a delay queue with one second as time period and after using all the slots within one second, none should be available:
public void givenDelayQueue_whenReachLimit_thenBlocked() {
int slots = 50;
ExecutorService executorService = Executors.newFixedThreadPool(slots);
DelayQueueUsingTimedSemaphore delayQueue
= new DelayQueueUsingTimedSemaphore(1, slots);
IntStream.range(0, slots)
.forEach(user -> executorService.execute(delayQueue::tryAdd));
executorService.shutdown();
assertEquals(0, delayQueue.availableSlots());
assertFalse(delayQueue.tryAdd());
}
But after sleeping for some time, the semaphore should reset and release the permits:
@Test
public void givenDelayQueue_whenTimePass_thenSlotsAvailable() throws InterruptedException {
int slots = 50;
ExecutorService executorService = Executors.newFixedThreadPool(slots);
DelayQueueUsingTimedSemaphore delayQueue = new DelayQueueUsingTimedSemaphore(1, slots);
IntStream.range(0, slots)
.forEach(user -> executorService.execute(delayQueue::tryAdd));
executorService.shutdown();
assertEquals(0, delayQueue.availableSlots());
Thread.sleep(1000);
assertTrue(delayQueue.availableSlots() > 0);
assertTrue(delayQueue.tryAdd());
}
4. Semaphore vs. Mutex
In the following example, we’ll use a simple binary semaphore to build a counter:
class CounterUsingMutex {
private Semaphore mutex;
private int count;
CounterUsingMutex() {
mutex = new Semaphore(1);
count = 0;
}
void increase() throws InterruptedException {
mutex.acquire();
this.count = this.count + 1;
Thread.sleep(1000);
mutex.release();
}
int getCount() {
return this.count;
}
boolean hasQueuedThreads() {
return mutex.hasQueuedThreads();
}
}
When a lot of threads try to access the counter at once, they’ll simply be blocked in a queue:
@Test
public void whenMutexAndMultipleThreads_thenBlocked()
throws InterruptedException {
int count = 5;
ExecutorService executorService
= Executors.newFixedThreadPool(count);
CounterUsingMutex counter = new CounterUsingMutex();
IntStream.range(0, count)
.forEach(user -> executorService.execute(() -> {
try {
counter.increase();
} catch (InterruptedException e) {
e.printStackTrace();
}
}));
executorService.shutdown();
assertTrue(counter.hasQueuedThreads());
}
When we wait, all threads will access the counter and no threads left in the queue:
@Test
public void givenMutexAndMultipleThreads_ThenDelay_thenCorrectCount()
throws InterruptedException {
int count = 5;
ExecutorService executorService
= Executors.newFixedThreadPool(count);
CounterUsingMutex counter = new CounterUsingMutex();
IntStream.range(0, count)
.forEach(user -> executorService.execute(() -> {
try {
counter.increase();
} catch (InterruptedException e) {
e.printStackTrace();
}
}));
executorService.shutdown();
assertTrue(counter.hasQueuedThreads());
Thread.sleep(5000);
assertFalse(counter.hasQueuedThreads());
assertEquals(count, counter.getCount());
}
5. Conclusion
As always, the full source code is available over on GitHub.