org.apache.hadoop.hbase.ChoreService Maven / Gradle / Ivy
Show all versions of hbase-common Show documentation
/**
*
* 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.hadoop.hbase;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.LinkedHashMap;
import java.util.Map.Entry;
import java.util.concurrent.ScheduledFuture;
import java.util.concurrent.ScheduledThreadPoolExecutor;
import java.util.concurrent.ThreadFactory;
import java.util.concurrent.atomic.AtomicInteger;
import com.google.common.annotations.VisibleForTesting;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.hbase.ScheduledChore.ChoreServicer;
import org.apache.hadoop.hbase.classification.InterfaceAudience;
import org.apache.hadoop.hbase.classification.InterfaceStability;
/**
* ChoreService is a service that can be used to schedule instances of {@link ScheduledChore} to run
* periodically while sharing threads. The ChoreService is backed by a
* {@link ScheduledThreadPoolExecutor} whose core pool size changes dynamically depending on the
* number of {@link ScheduledChore} scheduled. All of the threads in the core thread pool of the
* underlying {@link ScheduledThreadPoolExecutor} are set to be daemon threads.
*
* The ChoreService provides the ability to schedule, cancel, and trigger instances of
* {@link ScheduledChore}. The ChoreService also provides the ability to check on the status of
* scheduled chores. The number of threads used by the ChoreService changes based on the scheduling
* load and whether or not the scheduled chores are executing on time. As more chores are scheduled,
* there may be a need to increase the number of threads if it is noticed that chores are no longer
* meeting their scheduled start times. On the other hand, as chores are cancelled, an attempt is
* made to reduce the number of running threads to see if chores can still meet their start times
* with a smaller thread pool.
*
* When finished with a ChoreService it is good practice to call {@link ChoreService#shutdown()}.
* Calling this method ensures that all scheduled chores are cancelled and cleaned up properly.
*/
@InterfaceAudience.Public
@InterfaceStability.Stable
public class ChoreService implements ChoreServicer {
private static final Log LOG = LogFactory.getLog(ChoreService.class);
/**
* The minimum number of threads in the core pool of the underlying ScheduledThreadPoolExecutor
*/
@InterfaceAudience.Private
public final static int MIN_CORE_POOL_SIZE = 1;
/**
* This thread pool is used to schedule all of the Chores
*/
private final ScheduledThreadPoolExecutor scheduler;
/**
* Maps chores to their futures. Futures are used to control a chore's schedule
*/
private final HashMap> scheduledChores;
/**
* Maps chores to Booleans which indicate whether or not a chore has caused an increase in the
* core pool size of the ScheduledThreadPoolExecutor. Each chore should only be allowed to
* increase the core pool size by 1 (otherwise a single long running chore whose execution is
* longer than its period would be able to spawn too many threads).
*/
private final HashMap choresMissingStartTime;
/**
* The coreThreadPoolPrefix is the prefix that will be applied to all threads within the
* ScheduledThreadPoolExecutor. The prefix is typically related to the Server that the service is
* running on. The prefix is useful because it allows us to monitor how the thread pool of a
* particular service changes over time VIA thread dumps.
*/
private final String coreThreadPoolPrefix;
/**
*
* @param coreThreadPoolPrefix Prefix that will be applied to the Thread name of all threads
* spawned by this service
*/
@InterfaceAudience.Private
@VisibleForTesting
public ChoreService(final String coreThreadPoolPrefix) {
this(coreThreadPoolPrefix, MIN_CORE_POOL_SIZE, false);
}
/**
* @param coreThreadPoolPrefix Prefix that will be applied to the Thread name of all threads
* spawned by this service
* @param jitter Should chore service add some jitter for all of the scheduled chores. When set
* to true this will add -10% to 10% jitter.
*/
public ChoreService(final String coreThreadPoolPrefix, final boolean jitter) {
this(coreThreadPoolPrefix, MIN_CORE_POOL_SIZE, jitter);
}
/**
* @param coreThreadPoolPrefix Prefix that will be applied to the Thread name of all threads
* spawned by this service
* @param corePoolSize The initial size to set the core pool of the ScheduledThreadPoolExecutor
* to during initialization. The default size is 1, but specifying a larger size may be
* beneficial if you know that 1 thread will not be enough.
* @param jitter Should chore service add some jitter for all of the scheduled chores. When set
* to true this will add -10% to 10% jitter.
*/
public ChoreService(final String coreThreadPoolPrefix, int corePoolSize, boolean jitter) {
this.coreThreadPoolPrefix = coreThreadPoolPrefix;
if (corePoolSize < MIN_CORE_POOL_SIZE) {
corePoolSize = MIN_CORE_POOL_SIZE;
}
final ThreadFactory threadFactory = new ChoreServiceThreadFactory(coreThreadPoolPrefix);
if (jitter) {
scheduler = new JitterScheduledThreadPoolExecutorImpl(corePoolSize, threadFactory, 0.1);
} else {
scheduler = new ScheduledThreadPoolExecutor(corePoolSize, threadFactory);
}
scheduler.setRemoveOnCancelPolicy(true);
scheduledChores = new HashMap>();
choresMissingStartTime = new HashMap();
}
/**
* @param chore Chore to be scheduled. If the chore is already scheduled with another ChoreService
* instance, that schedule will be cancelled (i.e. a Chore can only ever be scheduled
* with a single ChoreService instance).
* @return true when the chore was successfully scheduled. false when the scheduling failed
* (typically occurs when a chore is scheduled during shutdown of service)
*/
public synchronized boolean scheduleChore(ScheduledChore chore) {
if (chore == null) {
return false;
}
try {
chore.setChoreServicer(this);
ScheduledFuture future =
scheduler.scheduleAtFixedRate(chore, chore.getInitialDelay(), chore.getPeriod(),
chore.getTimeUnit());
scheduledChores.put(chore, future);
return true;
} catch (Exception exception) {
if (LOG.isInfoEnabled()) {
LOG.info("Could not successfully schedule chore: " + chore.getName());
}
return false;
}
}
/**
* @param chore The Chore to be rescheduled. If the chore is not scheduled with this ChoreService
* yet then this call is equivalent to a call to scheduleChore.
*/
private synchronized void rescheduleChore(ScheduledChore chore) {
if (chore == null) return;
if (scheduledChores.containsKey(chore)) {
ScheduledFuture future = scheduledChores.get(chore);
future.cancel(false);
}
scheduleChore(chore);
}
@InterfaceAudience.Private
@Override
public synchronized void cancelChore(ScheduledChore chore) {
cancelChore(chore, true);
}
@InterfaceAudience.Private
@Override
public synchronized void cancelChore(ScheduledChore chore, boolean mayInterruptIfRunning) {
if (chore != null && scheduledChores.containsKey(chore)) {
ScheduledFuture future = scheduledChores.get(chore);
future.cancel(mayInterruptIfRunning);
scheduledChores.remove(chore);
// Removing a chore that was missing its start time means it may be possible
// to reduce the number of threads
if (choresMissingStartTime.containsKey(chore)) {
choresMissingStartTime.remove(chore);
requestCorePoolDecrease();
}
}
}
@InterfaceAudience.Private
@Override
public synchronized boolean isChoreScheduled(ScheduledChore chore) {
return chore != null && scheduledChores.containsKey(chore)
&& !scheduledChores.get(chore).isDone();
}
@InterfaceAudience.Private
@Override
public synchronized boolean triggerNow(ScheduledChore chore) {
if (chore == null) {
return false;
} else {
rescheduleChore(chore);
return true;
}
}
/**
* @return number of chores that this service currently has scheduled
*/
int getNumberOfScheduledChores() {
return scheduledChores.size();
}
/**
* @return number of chores that this service currently has scheduled that are missing their
* scheduled start time
*/
int getNumberOfChoresMissingStartTime() {
return choresMissingStartTime.size();
}
/**
* @return number of threads in the core pool of the underlying ScheduledThreadPoolExecutor
*/
int getCorePoolSize() {
return scheduler.getCorePoolSize();
}
/**
* Custom ThreadFactory used with the ScheduledThreadPoolExecutor so that all the threads are
* daemon threads, and thus, don't prevent the JVM from shutting down
*/
static class ChoreServiceThreadFactory implements ThreadFactory {
private final String threadPrefix;
private final static String THREAD_NAME_SUFFIX = "_ChoreService_";
private AtomicInteger threadNumber = new AtomicInteger(1);
/**
* @param threadPrefix The prefix given to all threads created by this factory
*/
public ChoreServiceThreadFactory(final String threadPrefix) {
this.threadPrefix = threadPrefix;
}
@Override
public Thread newThread(Runnable r) {
Thread thread =
new Thread(r, threadPrefix + THREAD_NAME_SUFFIX + threadNumber.getAndIncrement());
thread.setDaemon(true);
return thread;
}
}
/**
* Represents a request to increase the number of core pool threads. Typically a request
* originates from the fact that the current core pool size is not sufficient to service all of
* the currently running Chores
* @return true when the request to increase the core pool size succeeds
*/
private synchronized boolean requestCorePoolIncrease() {
// There is no point in creating more threads than scheduledChores.size since scheduled runs
// of the same chore cannot run concurrently (i.e. happen-before behavior is enforced
// amongst occurrences of the same chore).
if (scheduler.getCorePoolSize() < scheduledChores.size()) {
scheduler.setCorePoolSize(scheduler.getCorePoolSize() + 1);
printChoreServiceDetails("requestCorePoolIncrease");
return true;
}
return false;
}
/**
* Represents a request to decrease the number of core pool threads. Typically a request
* originates from the fact that the current core pool size is more than sufficient to service the
* running Chores.
*/
private synchronized void requestCorePoolDecrease() {
if (scheduler.getCorePoolSize() > MIN_CORE_POOL_SIZE) {
scheduler.setCorePoolSize(scheduler.getCorePoolSize() - 1);
printChoreServiceDetails("requestCorePoolDecrease");
}
}
@InterfaceAudience.Private
@Override
public synchronized void onChoreMissedStartTime(ScheduledChore chore) {
if (chore == null || !scheduledChores.containsKey(chore)) return;
// If the chore has not caused an increase in the size of the core thread pool then request an
// increase. This allows each chore missing its start time to increase the core pool size by
// at most 1.
if (!choresMissingStartTime.containsKey(chore) || !choresMissingStartTime.get(chore)) {
choresMissingStartTime.put(chore, requestCorePoolIncrease());
}
// Must reschedule the chore to prevent unnecessary delays of chores in the scheduler. If
// the chore is NOT rescheduled, future executions of this chore will be delayed more and
// more on each iteration. This hurts us because the ScheduledThreadPoolExecutor allocates
// idle threads to chores based on how delayed they are.
rescheduleChore(chore);
printChoreDetails("onChoreMissedStartTime", chore);
}
/**
* shutdown the service. Any chores that are scheduled for execution will be cancelled. Any chores
* in the middle of execution will be interrupted and shutdown. This service will be unusable
* after this method has been called (i.e. future scheduling attempts will fail).
*/
public synchronized void shutdown() {
scheduler.shutdownNow();
if (LOG.isInfoEnabled()) {
LOG.info("Chore service for: " + coreThreadPoolPrefix + " had " + scheduledChores.keySet()
+ " on shutdown");
}
cancelAllChores(true);
scheduledChores.clear();
choresMissingStartTime.clear();
}
/**
* @return true when the service is shutdown and thus cannot be used anymore
*/
public boolean isShutdown() {
return scheduler.isShutdown();
}
/**
* @return true when the service is shutdown and all threads have terminated
*/
public boolean isTerminated() {
return scheduler.isTerminated();
}
private void cancelAllChores(final boolean mayInterruptIfRunning) {
ArrayList choresToCancel = new ArrayList();
// Build list of chores to cancel so we can iterate through a set that won't change
// as chores are cancelled. If we tried to cancel each chore while iterating through
// keySet the results would be undefined because the keySet would be changing
for (ScheduledChore chore : scheduledChores.keySet()) {
choresToCancel.add(chore);
}
for (ScheduledChore chore : choresToCancel) {
cancelChore(chore, mayInterruptIfRunning);
}
choresToCancel.clear();
}
/**
* Prints a summary of important details about the chore. Used for debugging purposes
*/
private void printChoreDetails(final String header, ScheduledChore chore) {
LinkedHashMap output = new LinkedHashMap();
output.put(header, "");
output.put("Chore name: ", chore.getName());
output.put("Chore period: ", Integer.toString(chore.getPeriod()));
output.put("Chore timeBetweenRuns: ", Long.toString(chore.getTimeBetweenRuns()));
for (Entry entry : output.entrySet()) {
if (LOG.isTraceEnabled()) LOG.trace(entry.getKey() + entry.getValue());
}
}
/**
* Prints a summary of important details about the service. Used for debugging purposes
*/
private void printChoreServiceDetails(final String header) {
LinkedHashMap output = new LinkedHashMap();
output.put(header, "");
output.put("ChoreService corePoolSize: ", Integer.toString(getCorePoolSize()));
output.put("ChoreService scheduledChores: ", Integer.toString(getNumberOfScheduledChores()));
output.put("ChoreService missingStartTimeCount: ",
Integer.toString(getNumberOfChoresMissingStartTime()));
for (Entry entry : output.entrySet()) {
if (LOG.isTraceEnabled()) LOG.trace(entry.getKey() + entry.getValue());
}
}
}