/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.apache.commons.lang3.concurrent;

import org.apache.commons.lang3.Validate;

import java.util.concurrent.ScheduledExecutorService;
import java.util.concurrent.ScheduledFuture;
import java.util.concurrent.ScheduledThreadPoolExecutor;
import java.util.concurrent.TimeUnit;

A specialized semaphore implementation that provides a number of permits in a given time frame.

This class is similar to the java.util.concurrent.Semaphore class provided by the JDK in that it manages a configurable number of permits. Using the acquire() method a permit can be requested by a thread. However, there is an additional timing dimension: there is no release() method for freeing a permit, but all permits are automatically released at the end of a configurable time frame. If a thread calls acquire() and the available permits are already exhausted for this time frame, the thread is blocked. When the time frame ends all permits requested so far are restored, and blocking threads are waked up again, so that they can try to acquire a new permit. This basically means that in the specified time frame only the given number of operations is possible.

A use case for this class is to artificially limit the load produced by a process. As an example consider an application that issues database queries on a production system in a background process to gather statistical information. This background processing should not produce so much database load that the functionality and the performance of the production system are impacted. Here a TimedSemaphore could be installed to guarantee that only a given number of database queries are issued per second.

A thread class for performing database queries could look as follows:

public class StatisticsThread extends Thread {
    // The semaphore for limiting database load.
    private final TimedSemaphore semaphore;
    // Create an instance and set the semaphore
    public StatisticsThread(TimedSemaphore timedSemaphore) {
        semaphore = timedSemaphore;
    }
    // Gather statistics
    public void run() {
        try {
            while(true) {
                semaphore.acquire();   // limit database load
                performQuery();        // issue a query
            }
        } catch(InterruptedException) {
            // fall through
        }
    }
    ...
}

The following code fragment shows how a TimedSemaphore is created that allows only 10 operations per second and passed to the statistics thread:

TimedSemaphore sem = new TimedSemaphore(1, TimeUnit.SECOND, 10);
StatisticsThread thread = new StatisticsThread(sem);
thread.start();

When creating an instance the time period for the semaphore must be specified. TimedSemaphore uses an executor service with a corresponding period to monitor this interval. The ScheduledExecutorService to be used for this purpose can be provided at construction time. Alternatively the class creates an internal executor service.

Client code that uses TimedSemaphore has to call the acquire() method in each processing step. TimedSemaphore keeps track of the number of invocations of the acquire() method and blocks the calling thread if the counter exceeds the limit specified. When the timer signals the end of the time period the counter is reset and all waiting threads are released. Then another cycle can start.

An alternative to acquire() is the tryAcquire() method. This method checks whether the semaphore is under the specified limit and increases the internal counter if this is the case. The return value is then true, and the calling thread can continue with its action. If the semaphore is already at its limit, tryAcquire() immediately returns false without blocking; the calling thread must then abort its action. This usage scenario prevents blocking of threads.

It is possible to modify the limit at any time using the setLimit(int) method. This is useful if the load produced by an operation has to be adapted dynamically. In the example scenario with the thread collecting statistics it may make sense to specify a low limit during day time while allowing a higher load in the night time. Reducing the limit takes effect immediately by blocking incoming callers. If the limit is increased, waiting threads are not released immediately, but wake up when the timer runs out. Then, in the next period more processing steps can be performed without blocking. By setting the limit to 0 the semaphore can be switched off: in this mode the acquire() method never blocks, but lets all callers pass directly.

When the TimedSemaphore is no more needed its shutdown() method should be called. This causes the periodic task that monitors the time interval to be canceled. If the ScheduledExecutorService has been created by the semaphore at construction time, it is also shut down. resources. After that acquire() must not be called any more.

Since:3.0
/** * <p> * A specialized <em>semaphore</em> implementation that provides a number of * permits in a given time frame. * </p> * <p> * This class is similar to the {@code java.util.concurrent.Semaphore} class * provided by the JDK in that it manages a configurable number of permits. * Using the {@link #acquire()} method a permit can be requested by a thread. * However, there is an additional timing dimension: there is no {@code * release()} method for freeing a permit, but all permits are automatically * released at the end of a configurable time frame. If a thread calls * {@link #acquire()} and the available permits are already exhausted for this * time frame, the thread is blocked. When the time frame ends all permits * requested so far are restored, and blocking threads are waked up again, so * that they can try to acquire a new permit. This basically means that in the * specified time frame only the given number of operations is possible. * </p> * <p> * A use case for this class is to artificially limit the load produced by a * process. As an example consider an application that issues database queries * on a production system in a background process to gather statistical * information. This background processing should not produce so much database * load that the functionality and the performance of the production system are * impacted. Here a {@code TimedSemaphore} could be installed to guarantee that * only a given number of database queries are issued per second. * </p> * <p> * A thread class for performing database queries could look as follows: * </p> * * <pre> * public class StatisticsThread extends Thread { * // The semaphore for limiting database load. * private final TimedSemaphore semaphore; * // Create an instance and set the semaphore * public StatisticsThread(TimedSemaphore timedSemaphore) { * semaphore = timedSemaphore; * } * // Gather statistics * public void run() { * try { * while(true) { * semaphore.acquire(); // limit database load * performQuery(); // issue a query * } * } catch(InterruptedException) { * // fall through * } * } * ... * } * </pre> * * <p> * The following code fragment shows how a {@code TimedSemaphore} is created * that allows only 10 operations per second and passed to the statistics * thread: * </p> * * <pre> * TimedSemaphore sem = new TimedSemaphore(1, TimeUnit.SECOND, 10); * StatisticsThread thread = new StatisticsThread(sem); * thread.start(); * </pre> * * <p> * When creating an instance the time period for the semaphore must be * specified. {@code TimedSemaphore} uses an executor service with a * corresponding period to monitor this interval. The {@code * ScheduledExecutorService} to be used for this purpose can be provided at * construction time. Alternatively the class creates an internal executor * service. * </p> * <p> * Client code that uses {@code TimedSemaphore} has to call the * {@link #acquire()} method in each processing step. {@code TimedSemaphore} * keeps track of the number of invocations of the {@link #acquire()} method and * blocks the calling thread if the counter exceeds the limit specified. When * the timer signals the end of the time period the counter is reset and all * waiting threads are released. Then another cycle can start. * </p> * <p> * An alternative to {@code acquire()} is the {@link #tryAcquire()} method. This * method checks whether the semaphore is under the specified limit and * increases the internal counter if this is the case. The return value is then * <strong>true</strong>, and the calling thread can continue with its action. * If the semaphore is already at its limit, {@code tryAcquire()} immediately * returns <strong>false</strong> without blocking; the calling thread must * then abort its action. This usage scenario prevents blocking of threads. * </p> * <p> * It is possible to modify the limit at any time using the * {@link #setLimit(int)} method. This is useful if the load produced by an * operation has to be adapted dynamically. In the example scenario with the * thread collecting statistics it may make sense to specify a low limit during * day time while allowing a higher load in the night time. Reducing the limit * takes effect immediately by blocking incoming callers. If the limit is * increased, waiting threads are not released immediately, but wake up when the * timer runs out. Then, in the next period more processing steps can be * performed without blocking. By setting the limit to 0 the semaphore can be * switched off: in this mode the {@link #acquire()} method never blocks, but * lets all callers pass directly. * </p> * <p> * When the {@code TimedSemaphore} is no more needed its {@link #shutdown()} * method should be called. This causes the periodic task that monitors the time * interval to be canceled. If the {@code ScheduledExecutorService} has been * created by the semaphore at construction time, it is also shut down. * resources. After that {@link #acquire()} must not be called any more. * </p> * * @since 3.0 */
public class TimedSemaphore {
Constant for a value representing no limit. If the limit is set to a value less or equal this constant, the TimedSemaphore will be effectively switched off.
/** * Constant for a value representing no limit. If the limit is set to a * value less or equal this constant, the {@code TimedSemaphore} will be * effectively switched off. */
public static final int NO_LIMIT = 0;
Constant for the thread pool size for the executor.
/** Constant for the thread pool size for the executor. */
private static final int THREAD_POOL_SIZE = 1;
The executor service for managing the timer thread.
/** The executor service for managing the timer thread. */
private final ScheduledExecutorService executorService;
Stores the period for this timed semaphore.
/** Stores the period for this timed semaphore. */
private final long period;
The time unit for the period.
/** The time unit for the period. */
private final TimeUnit unit;
A flag whether the executor service was created by this object.
/** A flag whether the executor service was created by this object. */
private final boolean ownExecutor;
A future object representing the timer task.
/** A future object representing the timer task. */
private ScheduledFuture<?> task; // @GuardedBy("this")
Stores the total number of invocations of the acquire() method.
/** Stores the total number of invocations of the acquire() method. */
private long totalAcquireCount; // @GuardedBy("this")
The counter for the periods. This counter is increased every time a period ends.
/** * The counter for the periods. This counter is increased every time a * period ends. */
private long periodCount; // @GuardedBy("this")
The limit.
/** The limit. */
private int limit; // @GuardedBy("this")
The current counter.
/** The current counter. */
private int acquireCount; // @GuardedBy("this")
The number of invocations of acquire() in the last period.
/** The number of invocations of acquire() in the last period. */
private int lastCallsPerPeriod; // @GuardedBy("this")
A flag whether shutdown() was called.
/** A flag whether shutdown() was called. */
private boolean shutdown; // @GuardedBy("this")
Creates a new instance of TimedSemaphore and initializes it with the given time period and the limit.
Params:
  • timePeriod – the time period
  • timeUnit – the unit for the period
  • limit – the limit for the semaphore
Throws:
/** * Creates a new instance of {@link TimedSemaphore} and initializes it with * the given time period and the limit. * * @param timePeriod the time period * @param timeUnit the unit for the period * @param limit the limit for the semaphore * @throws IllegalArgumentException if the period is less or equals 0 */
public TimedSemaphore(final long timePeriod, final TimeUnit timeUnit, final int limit) { this(null, timePeriod, timeUnit, limit); }
Creates a new instance of TimedSemaphore and initializes it with an executor service, the given time period, and the limit. The executor service will be used for creating a periodic task for monitoring the time period. It can be null, then a default service will be created.
Params:
  • service – the executor service
  • timePeriod – the time period
  • timeUnit – the unit for the period
  • limit – the limit for the semaphore
Throws:
/** * Creates a new instance of {@link TimedSemaphore} and initializes it with * an executor service, the given time period, and the limit. The executor * service will be used for creating a periodic task for monitoring the time * period. It can be <b>null</b>, then a default service will be created. * * @param service the executor service * @param timePeriod the time period * @param timeUnit the unit for the period * @param limit the limit for the semaphore * @throws IllegalArgumentException if the period is less or equals 0 */
public TimedSemaphore(final ScheduledExecutorService service, final long timePeriod, final TimeUnit timeUnit, final int limit) { Validate.inclusiveBetween(1, Long.MAX_VALUE, timePeriod, "Time period must be greater than 0!"); period = timePeriod; unit = timeUnit; if (service != null) { executorService = service; ownExecutor = false; } else { final ScheduledThreadPoolExecutor s = new ScheduledThreadPoolExecutor( THREAD_POOL_SIZE); s.setContinueExistingPeriodicTasksAfterShutdownPolicy(false); s.setExecuteExistingDelayedTasksAfterShutdownPolicy(false); executorService = s; ownExecutor = true; } setLimit(limit); }
Returns the limit enforced by this semaphore. The limit determines how many invocations of acquire() are allowed within the monitored period.
Returns:the limit
/** * Returns the limit enforced by this semaphore. The limit determines how * many invocations of {@link #acquire()} are allowed within the monitored * period. * * @return the limit */
public final synchronized int getLimit() { return limit; }
Sets the limit. This is the number of times the acquire() method can be called within the time period specified. If this limit is reached, further invocations of acquire() will block. Setting the limit to a value <= NO_LIMIT will cause the limit to be disabled, i.e. an arbitrary number ofacquire() invocations is allowed in the time period.
Params:
  • limit – the limit
/** * Sets the limit. This is the number of times the {@link #acquire()} method * can be called within the time period specified. If this limit is reached, * further invocations of {@link #acquire()} will block. Setting the limit * to a value &lt;= {@link #NO_LIMIT} will cause the limit to be disabled, * i.e. an arbitrary number of{@link #acquire()} invocations is allowed in * the time period. * * @param limit the limit */
public final synchronized void setLimit(final int limit) { this.limit = limit; }
Initializes a shutdown. After that the object cannot be used any more. This method can be invoked an arbitrary number of times. All invocations after the first one do not have any effect.
/** * Initializes a shutdown. After that the object cannot be used any more. * This method can be invoked an arbitrary number of times. All invocations * after the first one do not have any effect. */
public synchronized void shutdown() { if (!shutdown) { if (ownExecutor) { // if the executor was created by this instance, it has // to be shutdown getExecutorService().shutdownNow(); } if (task != null) { task.cancel(false); } shutdown = true; } }
Tests whether the shutdown() method has been called on this object. If this method returns true, this instance cannot be used any longer.
Returns:a flag whether a shutdown has been performed
/** * Tests whether the {@link #shutdown()} method has been called on this * object. If this method returns <b>true</b>, this instance cannot be used * any longer. * * @return a flag whether a shutdown has been performed */
public synchronized boolean isShutdown() { return shutdown; }
Acquires a permit from this semaphore. This method will block if the limit for the current period has already been reached. If shutdown() has already been invoked, calling this method will cause an exception. The very first call of this method starts the timer task which monitors the time period set for this TimedSemaphore. From now on the semaphore is active.
Throws:
/** * Acquires a permit from this semaphore. This method will block if * the limit for the current period has already been reached. If * {@link #shutdown()} has already been invoked, calling this method will * cause an exception. The very first call of this method starts the timer * task which monitors the time period set for this {@code TimedSemaphore}. * From now on the semaphore is active. * * @throws InterruptedException if the thread gets interrupted * @throws IllegalStateException if this semaphore is already shut down */
public synchronized void acquire() throws InterruptedException { prepareAcquire(); boolean canPass; do { canPass = acquirePermit(); if (!canPass) { wait(); } } while (!canPass); }
Tries to acquire a permit from this semaphore. If the limit of this semaphore has not yet been reached, a permit is acquired, and this method returns true. Otherwise, this method returns immediately with the result false.
Throws:
Returns:true if a permit could be acquired; false otherwise
Since:3.5
/** * Tries to acquire a permit from this semaphore. If the limit of this semaphore has * not yet been reached, a permit is acquired, and this method returns * <strong>true</strong>. Otherwise, this method returns immediately with the result * <strong>false</strong>. * * @return <strong>true</strong> if a permit could be acquired; <strong>false</strong> * otherwise * @throws IllegalStateException if this semaphore is already shut down * @since 3.5 */
public synchronized boolean tryAcquire() { prepareAcquire(); return acquirePermit(); }
Returns the number of (successful) acquire invocations during the last period. This is the number of times the acquire() method was called without blocking. This can be useful for testing or debugging purposes or to determine a meaningful threshold value. If a limit is set, the value returned by this method won't be greater than this limit.
Returns:the number of non-blocking invocations of the acquire() method
/** * Returns the number of (successful) acquire invocations during the last * period. This is the number of times the {@link #acquire()} method was * called without blocking. This can be useful for testing or debugging * purposes or to determine a meaningful threshold value. If a limit is set, * the value returned by this method won't be greater than this limit. * * @return the number of non-blocking invocations of the {@link #acquire()} * method */
public synchronized int getLastAcquiresPerPeriod() { return lastCallsPerPeriod; }
Returns the number of invocations of the acquire() method for the current period. This may be useful for testing or debugging purposes.
Returns:the current number of acquire() invocations
/** * Returns the number of invocations of the {@link #acquire()} method for * the current period. This may be useful for testing or debugging purposes. * * @return the current number of {@link #acquire()} invocations */
public synchronized int getAcquireCount() { return acquireCount; }
Returns the number of calls to the acquire() method that can still be performed in the current period without blocking. This method can give an indication whether it is safe to call the acquire() method without risking to be suspended. However, there is no guarantee that a subsequent call to acquire() actually is not-blocking because in the mean time other threads may have invoked the semaphore.
Returns:the current number of available acquire() calls in the current period
/** * Returns the number of calls to the {@link #acquire()} method that can * still be performed in the current period without blocking. This method * can give an indication whether it is safe to call the {@link #acquire()} * method without risking to be suspended. However, there is no guarantee * that a subsequent call to {@link #acquire()} actually is not-blocking * because in the mean time other threads may have invoked the semaphore. * * @return the current number of available {@link #acquire()} calls in the * current period */
public synchronized int getAvailablePermits() { return getLimit() - getAcquireCount(); }
Returns the average number of successful (i.e. non-blocking) acquire() invocations for the entire life-time of this TimedSemaphore. This method can be used for instance for statistical calculations.
Returns:the average number of acquire() invocations per time unit
/** * Returns the average number of successful (i.e. non-blocking) * {@link #acquire()} invocations for the entire life-time of this {@code * TimedSemaphore}. This method can be used for instance for statistical * calculations. * * @return the average number of {@link #acquire()} invocations per time * unit */
public synchronized double getAverageCallsPerPeriod() { return periodCount == 0 ? 0 : (double) totalAcquireCount / (double) periodCount; }
Returns the time period. This is the time monitored by this semaphore. Only a given number of invocations of the acquire() method is possible in this period.
Returns:the time period
/** * Returns the time period. This is the time monitored by this semaphore. * Only a given number of invocations of the {@link #acquire()} method is * possible in this period. * * @return the time period */
public long getPeriod() { return period; }
Returns the time unit. This is the unit used by getPeriod().
Returns:the time unit
/** * Returns the time unit. This is the unit used by {@link #getPeriod()}. * * @return the time unit */
public TimeUnit getUnit() { return unit; }
Returns the executor service used by this instance.
Returns:the executor service
/** * Returns the executor service used by this instance. * * @return the executor service */
protected ScheduledExecutorService getExecutorService() { return executorService; }
Starts the timer. This method is called when acquire() is called for the first time. It schedules a task to be executed at fixed rate to monitor the time period specified.
Returns:a future object representing the task scheduled
/** * Starts the timer. This method is called when {@link #acquire()} is called * for the first time. It schedules a task to be executed at fixed rate to * monitor the time period specified. * * @return a future object representing the task scheduled */
protected ScheduledFuture<?> startTimer() { return getExecutorService().scheduleAtFixedRate(new Runnable() { @Override public void run() { endOfPeriod(); } }, getPeriod(), getPeriod(), getUnit()); }
The current time period is finished. This method is called by the timer used internally to monitor the time period. It resets the counter and releases the threads waiting for this barrier.
/** * The current time period is finished. This method is called by the timer * used internally to monitor the time period. It resets the counter and * releases the threads waiting for this barrier. */
synchronized void endOfPeriod() { lastCallsPerPeriod = acquireCount; totalAcquireCount += acquireCount; periodCount++; acquireCount = 0; notifyAll(); }
Prepares an acquire operation. Checks for the current state and starts the internal timer if necessary. This method must be called with the lock of this object held.
/** * Prepares an acquire operation. Checks for the current state and starts the internal * timer if necessary. This method must be called with the lock of this object held. */
private void prepareAcquire() { if (isShutdown()) { throw new IllegalStateException("TimedSemaphore is shut down!"); } if (task == null) { task = startTimer(); } }
Internal helper method for acquiring a permit. This method checks whether currently a permit can be acquired and - if so - increases the internal counter. The return value indicates whether a permit could be acquired. This method must be called with the lock of this object held.
Returns:a flag whether a permit could be acquired
/** * Internal helper method for acquiring a permit. This method checks whether currently * a permit can be acquired and - if so - increases the internal counter. The return * value indicates whether a permit could be acquired. This method must be called with * the lock of this object held. * * @return a flag whether a permit could be acquired */
private boolean acquirePermit() { if (getLimit() <= NO_LIMIT || acquireCount < getLimit()) { acquireCount++; return true; } return false; } }