zookeeper 实现分布式队列

zookeeper是一种高性能，可扩展的服务。其本身提供了一致性保证，特点如下：

（1）顺序一致性：客户端的更新顺序与它们被发送的顺序一致。

（2）原子性：更新操作要么成功，要么失败，没有第三种结果。

（3）单系统镜像：无论客户端连接到哪一个服务器，他都将看到相同的zookeeper视图。

（4）可靠性：一旦一个更新操作被应用，那么在客户端再次更新之前，其值不会再改变。

zookeeper实现分布式队列

基于zookeeper 提供的一致性保证，我们可以实现一个分布式队列DistributedQueue。

主要的思路如下：

（1）首先DistributedQueue 的构造器中要传入
ZooKeeper：连接服务端的核心类；
dir：表示znode组的结点
acl：每个结点的访问控制列表

（2）所谓的入队，也就是在dir组下新建一个znode结点，不过结点类型是CreateMode.PERSISTENT_SEQUENTIAL，这就表明了新建结点的值根据序号是有一定顺序性的。其核心实现就是：
zookeeper.create(dir+"/"+prefix, data, acl, CreateMode.PERSISTENT_SEQUENTIAL);

（3）所谓的出队，其实就是取出所有的子结点，然后用treeMap维护有序性，然后将最小的结点出队(也就是在服务端删除结点最小的结点)

（4）该分布式队列的实现严重依赖于zookeeper的一致性保证。

具体的实现代码如下：

package com.zookeeper.lyt.queue;

import org.apache.zookeeper.*;
import org.apache.zookeeper.data.ACL;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import java.util.List;
import java.util.NoSuchElementException;
import java.util.TreeMap;
import java.util.concurrent.CountDownLatch;

/**
 * 
 * A protocol to implement a distributed queue.
 * 利用 zookeeper 构建分布式 队列
 *
 * 其依据的就是 zookeeper的一致性:顺序一致性,原子性,单系统镜像,可靠性.
 * 
 */
public class DistributedQueue {
    private static final Logger LOG = LoggerFactory.getLogger(DistributedQueue.class);

    private final String dir;

    private ZooKeeper zookeeper;
    private List acl = ZooDefs.Ids.OPEN_ACL_UNSAFE;

    private final String prefix = "qn-";


    public DistributedQueue(ZooKeeper zookeeper, String dir, List acl){
        this.dir = dir;

        if(acl != null){
            this.acl = acl;
        }
        this.zookeeper = zookeeper;

    }



    /**
     * Returns a Map of the children, ordered by id.
     *
     * 获取children 的Map集合, 根据ID来排序
     *
     * @param watcher optional watcher on getChildren() operation.
     * @return map from id to child name for all children
     */
    private TreeMap orderedChildren(Watcher watcher) throws KeeperException, InterruptedException {
        TreeMap orderedChildren = new TreeMap();

        List childNames = null;
        try{
            childNames = zookeeper.getChildren(dir, watcher);
        }catch (KeeperException.NoNodeException e){
            throw e;
        }

        for(String childName : childNames){
            try{
                //Check format
                if(!childName.regionMatches(0, prefix, 0, prefix.length())){
                    LOG.warn("Found child node with improper name: " + childName);
                    continue;
                }
                String suffix = childName.substring(prefix.length());
                Long childId = new Long(suffix);
                orderedChildren.put(childId,childName);
            }catch(NumberFormatException e){
                LOG.warn("Found child node with improper format : " + childName + " " + e,e);
            }
        }

        return orderedChildren;
    }

    /**
     * Find the smallest child node.
     *
     * 找到最小的结点
     * @return The name of the smallest child node.
     */
    private String smallestChildName() throws KeeperException, InterruptedException {
        long minId = Long.MAX_VALUE;
        String minName = "";

        List childNames = null;

        try{
            childNames = zookeeper.getChildren(dir, false);
        }catch(KeeperException.NoNodeException e){
            LOG.warn("Caught: " +e,e);
            return null;
        }

        for(String childName : childNames){
            try{
                //Check format
                if(!childName.regionMatches(0, prefix, 0, prefix.length())){
                    LOG.warn("Found child node with improper name: " + childName);
                    continue;
                }
                String suffix = childName.substring(prefix.length());
                long childId = Long.parseLong(suffix);
                if(childId < minId){
                    minId = childId;
                    minName = childName;
                }
            }catch(NumberFormatException e){
                LOG.warn("Found child node with improper format : " + childName + " " + e,e);
            }
        }


        if(minId < Long.MAX_VALUE){
            return minName;
        }else{
            return null;
        }
    }

    /**
     * Return the head of the queue without modifying the queue.
     * 返回队定的元素
     * @return the data at the head of the queue.
     * @throws NoSuchElementException
     * @throws KeeperException
     * @throws InterruptedException
     */
    public byte[] element() throws NoSuchElementException, KeeperException, InterruptedException {
        TreeMap orderedChildren;

        // element, take, and remove follow the same pattern.
        // We want to return the child node with the smallest sequence number.
        // Since other clients are remove()ing and take()ing nodes concurrently, 
        // the child with the smallest sequence number in orderedChildren might be gone by the time we check.
        // We don't call getChildren again until we have tried the rest of the nodes in sequence order.
        while(true){
            try{
                orderedChildren = orderedChildren(null);
            }catch(KeeperException.NoNodeException e){
                throw new NoSuchElementException();
            }
            if(orderedChildren.size() == 0 ) throw new NoSuchElementException();

            for(String headNode : orderedChildren.values()){
                if(headNode != null){
                    try{
                        return zookeeper.getData(dir+"/"+headNode, false, null);
                    }catch(KeeperException.NoNodeException e){
                        //Another client removed the node first, try next
                    }
                }
            }

        }
    }


    /**
     * Attempts to remove the head of the queue and return it.
     * @return The former head of the queue
     * @throws NoSuchElementException
     * @throws KeeperException
     * @throws InterruptedException
     */
    public byte[] remove() throws NoSuchElementException, KeeperException, InterruptedException {
        TreeMap orderedChildren;
        // Same as for element.  Should refactor this.
        while(true){
            try{
                orderedChildren = orderedChildren(null);
            }catch(KeeperException.NoNodeException e){
                throw new NoSuchElementException();
            }
            if(orderedChildren.size() == 0) throw new NoSuchElementException();

            for(String headNode : orderedChildren.values()){
                String path = dir +"/"+headNode;
                try{
                    byte[] data = zookeeper.getData(path, false, null);
                    zookeeper.delete(path, -1);
                    return data;
                }catch(KeeperException.NoNodeException e){
                    // Another client deleted the node first.
                }
            }

        }
    }

    private class LatchChildWatcher implements Watcher {

        CountDownLatch latch;

        public LatchChildWatcher(){
            latch = new CountDownLatch(1);
        }

        public void process(WatchedEvent event){
            LOG.debug("Watcher fired on path: " + event.getPath() + " state: " + 
                    event.getState() + " type " + event.getType());
            latch.countDown();
        }
        public void await() throws InterruptedException {
            latch.await();
        }
    }

    /**
     * Removes the head of the queue and returns it, blocks until it succeeds.
     * @return The former head of the queue
     * @throws NoSuchElementException
     * @throws KeeperException
     * @throws InterruptedException
     */
    public byte[] take() throws KeeperException, InterruptedException {
        TreeMap orderedChildren;
        // Same as for element.  Should refactor this.
        while(true){
            LatchChildWatcher childWatcher = new LatchChildWatcher();
            try{
                orderedChildren = orderedChildren(childWatcher);
            }catch(KeeperException.NoNodeException e){
                zookeeper.create(dir, new byte[0], acl, CreateMode.PERSISTENT);
                continue;
            }
            if(orderedChildren.size() == 0){
                childWatcher.await();
                continue;
            }

            for(String headNode : orderedChildren.values()){
                String path = dir +"/"+headNode;
                try{
                    byte[] data = zookeeper.getData(path, false, null);
                    zookeeper.delete(path, -1);
                    return data;
                }catch(KeeperException.NoNodeException e){
                    // Another client deleted the node first.
                }
            }
        }
    }

    /**
     * Inserts data into queue.
     * @param data
     * @return true if data was successfully added
     */
    public boolean offer(byte[] data) throws KeeperException, InterruptedException{
        for(;;){
            try{
                zookeeper.create(dir+"/"+prefix, data, acl, CreateMode.PERSISTENT_SEQUENTIAL);
                return true;
            }catch(KeeperException.NoNodeException e){
                zookeeper.create(dir, new byte[0], acl, CreateMode.PERSISTENT);
            }
        }

    }

    /**
     * Returns the data at the first element of the queue, or null if the queue is empty.
     * @return data at the first element of the queue, or null.
     * @throws KeeperException
     * @throws InterruptedException
     */
    public byte[] peek() throws KeeperException, InterruptedException{
        try{
            return element();
        }catch(NoSuchElementException e){
            return null;
        }
    }


    /**
     * Attempts to remove the head of the queue and return it. Returns null if the queue is empty.
     * @return Head of the queue or null.
     * @throws KeeperException
     * @throws InterruptedException
     */
    public byte[] poll() throws KeeperException, InterruptedException {
        try{
            return remove();
        }catch(NoSuchElementException e){
            return null;
        }
    }
}