/*
* Copyright (c) 2014 Brocade Communications Systems, Inc. and others. All rights reserved.
*
* This program and the accompanying materials are made available under the
* terms of the Eclipse Public License v1.0 which accompanies this distribution,
* and is available at http://www.eclipse.org/legal/epl-v10.html
*/
package org.opendaylight.yangtools.util.concurrent;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.TimeUnit;
/**
* Factory methods for creating {@link ExecutorService} instances with specific configurations.
* @author Thomas Pantelis
*/
public final class SpecialExecutors {
private SpecialExecutors() {
}
/**
* Creates an ExecutorService with a specified bounded queue capacity that favors creating new
* threads over queuing, as the former is faster, so threads will only be reused when the thread
* limit is exceeded and tasks are queued. If the maximum queue capacity is reached, subsequent
* tasks will be rejected.
*
* <p>For example, if the maximum number of threads is 100 and 100 short-lived tasks are submitted
* within say 10 seconds, then 100 threads will be created and used - previously constructed
* idle threads will not be reused. This provides the fastest execution of the 100 tasks at the
* expense of memory and thread resource overhead. Therefore it is advisable to specify a
* relatively small thread limit (probably no more than 50).
*
* <p>Threads that have not been used for 15 seconds are terminated and removed from the pool.
* Thus, a pool that remains idle for long enough will not consume any resources.
*
* <p>If you need an executor with less memory and thread resource overhead where slower execution
* time is acceptable, consider using {@link #newBoundedCachedThreadPool }.
*
* @param maximumPoolSize
* the maximum number of threads to allow in the pool. Threads will terminate after
* being idle for 15 seconds.
* @param maximumQueueSize
* the capacity of the queue.
* @param threadPrefix
* the name prefix for threads created by this executor.
* @return a new ExecutorService with the specified configuration.
*/
public static ExecutorService newBoundedFastThreadPool( int maximumPoolSize,
int maximumQueueSize, String threadPrefix ) {
return new FastThreadPoolExecutor( maximumPoolSize, maximumQueueSize, threadPrefix );
}
/**
* Creates an ExecutorService similar to {@link #newBoundedFastThreadPool } except that it
* handles rejected tasks by running them in the same thread as the caller. Therefore if the
* queue is full, the caller submitting the task will be blocked until the task completes. In
* this manner, tasks are never rejected.
*
* @param maximumPoolSize
* the maximum number of threads to allow in the pool. Threads will terminate after
* being idle for 15 seconds.
* @param maximumQueueSize
* the capacity of the queue.
* @param threadPrefix
* the name prefix for threads created by this executor.
* @return a new ExecutorService with the specified configuration.
*/
public static ExecutorService newBlockingBoundedFastThreadPool( int maximumPoolSize,
int maximumQueueSize, String threadPrefix ) {
FastThreadPoolExecutor executor =
new FastThreadPoolExecutor( maximumPoolSize, maximumQueueSize, threadPrefix );
executor.setRejectedExecutionHandler( CountingRejectedExecutionHandler.newCallerRunsPolicy() );
return executor;
}
/**
* Creates an ExecutorService with a specified bounded queue capacity that favors reusing
* previously constructed threads, when they are available, over creating new threads. When a
* task is submitted, if no existing thread is available, a new thread will be created and added
* to the pool. If there is an existing idle thread available, the task will be handed to that
* thread to execute. If the specified maximum thread limit is reached, subsequent tasks will be
* queued and will execute as threads become available. If the maximum queue capacity is
* reached, subsequent tasks will be rejected.
*
* <p>Threads that have not been used for sixty seconds are terminated and removed from the pool.
* Thus, a pool that remains idle for long enough will not consume any resources.
*
* <p>By reusing threads when possible, this executor optimizes for reduced memory and thread
* resource overhead at the expense of execution time.
*
* <p>If you need an executor with faster execution time where increased memory and thread resource
* overhead is acceptable, consider using {@link #newBoundedFastThreadPool }.
*
* @param maximumPoolSize
* the maximum number of threads to allow in the pool. Threads will terminate after
* being idle for 60 seconds.
* @param maximumQueueSize
* the capacity of the queue.
* @param threadPrefix
* the name prefix for threads created by this executor.
* @return a new ExecutorService with the specified configuration.
*/
public static ExecutorService newBoundedCachedThreadPool( int maximumPoolSize,
int maximumQueueSize, String threadPrefix ) {
return new CachedThreadPoolExecutor( maximumPoolSize, maximumQueueSize, threadPrefix );
}
/**
* Creates an ExecutorService similar to {@link #newBoundedCachedThreadPool } except that it
* handles rejected tasks by running them in the same thread as the caller. Therefore if the
* queue is full, the caller submitting the task will be blocked until the task completes. In
* this manner, tasks are never rejected.
*
* @param maximumPoolSize
* the maximum number of threads to allow in the pool. Threads will terminate after
* being idle for 60 seconds.
* @param maximumQueueSize
* the capacity of the queue.
* @param threadPrefix
* the name prefix for threads created by this executor.
* @return a new ExecutorService with the specified configuration.
*/
public static ExecutorService newBlockingBoundedCachedThreadPool( int maximumPoolSize,
int maximumQueueSize, String threadPrefix ) {
CachedThreadPoolExecutor executor =
new CachedThreadPoolExecutor( maximumPoolSize, maximumQueueSize, threadPrefix );
executor.setRejectedExecutionHandler( CountingRejectedExecutionHandler.newCallerRunsPolicy() );
return executor;
}
/**
* Creates an ExecutorService that uses a single worker thread operating off a bounded queue
* with the specified capacity. Tasks are guaranteed to execute sequentially, and no more than
* one task will be active at any given time. If the maximum queue capacity is reached,
* subsequent tasks will be rejected.
*
* @param maximumQueueSize
* the capacity of the queue.
* @param threadPrefix
* the name prefix for the thread created by this executor.
* @return a new ExecutorService with the specified configuration.
*/
public static ExecutorService newBoundedSingleThreadExecutor( int maximumQueueSize,
String threadPrefix ) {
return new FastThreadPoolExecutor( 1, maximumQueueSize, Long.MAX_VALUE, TimeUnit.SECONDS,
threadPrefix );
}
}