BoneCP主要使用了下面几种第三方包：
1、Google Guava library The Guava project contains several of Google's core libraries that we rely on in our Java-based projects: collections, caching, primitives support, concurrency libraries, common annotations, string processing, I/O, and so forth.
2、SLF4J 日志管理
3、jsr166y Doug Lea写的一个并发包，代码量不多，已合并到JDK7的JUC包中。BoneCP中使用该包下的LinkedTransferQueue队列来保存Connection对象：
if (config.getMaxConnectionsPerPartition() == config.getMinConnectionsPerPartition()){
// if we have a pool that we don't want resized, make it even faster by ignoring
// the size constraints.
connectionHandles = queueLIFO ? new LIFOQueue<ConnectionHandle>() : new LinkedTransferQueue<ConnectionHandle>();
} else {
connectionHandles = queueLIFO ? new LIFOQueue<ConnectionHandle>(this.config.getMaxConnectionsPerPartition()) : new BoundedLinkedTransferQueue<ConnectionHandle>(this.config.getMaxConnectionsPerPartition());
}
这段神奇的代码还没研究明白用意，当设置每个partition中的最大连接数和最小连接数相等时，保存Connection对象的队列就使用无参构造函数，也就是队列的最大值为Integer.MAX_VALUE，如果不相等则队列的最大值为用户设置的最大值。
BoneCP只使用到jsr166y中的3个类，合成到它自己的jar里：
TransferQueue和LinkedTransferQueue
著名的Java并发编程大师Doug lea在JDK7的并发包里新增一个队列集合类LinkedTransferQueue，
TransferQueue接口继承自BlockingQueue接口，LinkedTransferQueue是TransferQueue的实现，它基于单向双端链表，是个无界链表，它的Node仅持有其下一个节点的引用，是一个典型FIFO队列，如下为保存无线的Node类实现：
static final class Node {
final boolean isData; // false if this is a request node
volatile Object item; // initially non-null if isData; CASed to match
volatile Node next;
volatile Thread waiter; // null until waiting，存放当前线程，当执行LockSupport.park(this)时，当前线程被阻塞，当执行LockSupport.unpark(node.waiter)时，该节点对应的线程将解除阻塞
// CAS methods for fields
final boolean casNext(Node cmp, Node val) {
return UNSAFE.compareAndSwapObject(this, nextOffset, cmp, val);
}
......
}
该类对链表的操作不加锁，而是采用第三方的sun.misc.Unsafe的CAS操作来保证同步：
private static final Unsafe UNSAFE = getUnsafe();
// CAS methods for fields
private boolean casTail(Node cmp, Node val) {
return UNSAFE.compareAndSwapObject(this, tailOffset, cmp, val);
}
private boolean casHead(Node cmp, Node val) {
return UNSAFE.compareAndSwapObject(this, headOffset, cmp, val);
}
private boolean casSweepVotes(int cmp, int val) {
return UNSAFE.compareAndSwapInt(this, sweepVotesOffset, cmp, val);
}
该类中操作链表的方法都调用下面方法来实现：
public E poll() {
return xfer(null, false, NOW, 0);
}
private E xfer(E e, boolean haveData, int how, long nanos) {
if (haveData && (e == null))
throw new NullPointerException();
Node s = null; // the node to append, if needed
retry: for (;;) { // restart on append race
for (Node h = head, p = h; p != null;) { // find & match first node
boolean isData = p.isData;
Object item = p.item;
if (item != p && (item != null) == isData) { // unmatched
if (isData == haveData) // can't match
break;
if (p.casItem(item, e)) { // match
for (Node q = p; q != h;) {
Node n = q.next; // update by 2 unless singleton
if (head == h && casHead(h, n == null? q : n)) {
h.forgetNext();
break;
} // advance and retry
if ((h = head) == null ||
(q = h.next) == null || !q.isMatched())
break; // unless slack < 2
}
LockSupport.unpark(p.waiter);
return this.<E>cast(item);
}
}
Node n = p.next;
p = (p != n) ? n : (h = head); // Use head if p offlist
}
if (how != NOW) { // No matches available
if (s == null)
s = new Node(e, haveData);
Node pred = tryAppend(s, haveData);
if (pred == null)
continue retry; // lost race vs opposite mode
if (how != ASYNC)
return awaitMatch(s, pred, e, (how == TIMED), nanos);
}
return e; // not waiting
}
} 四个特殊变量NOW、ASYNC、SYNC、TIMED在poll、put等方法调用efer方法时使用：
/*
* Possible values for "how" argument in xfer method.
*/
private static final int NOW = 0; // for untimed poll, tryTransfer，用于不带超时的poll和tryTransfer方法
private static final int ASYNC = 1; // for offer, put, add，用于offer，put和add方法
private static final int SYNC = 2; // for transfer, take，用于transfer和take方法
private static final int TIMED = 3; // for timed poll, tryTransfer，用于带超时的poll和tryTransfer方法
插入元素过程
LinkedTransferQueue插入元素，元素都插入到队尾，方法有下面几种：
//在队尾插入元素，由于队列是无界的所以不会阻塞
public void put(E e) {
xfer(e, true, ASYNC, 0);
}
//在队尾插入元素，永远返回true
public boolean offer(E e) {
xfer(e, true, ASYNC, 0);
return true;
}
//同上
public boolean offer(E e, long timeout, TimeUnit unit) {
xfer(e, true, ASYNC, 0);
return true;
}
//同上
public boolean add(E e) {
xfer(e, true, ASYNC, 0);
return true;
}
//若当前存在一个正在等待获取的消费者线程，即立刻移交之；否则，会插入当前元素e到队列尾部，并且等待进入阻塞状态，到有消费者线程取走该元素。
public void transfer(E e) throws InterruptedException {
if (xfer(e, true, SYNC, 0) != null) {
Thread.interrupted(); // failure possible only due to interrupt
throw new InterruptedException();
}
}
//若当前存在一个正在等待获取的消费者线程（使用take()或者poll()函数），使用该方法会即刻转移/传输对象元素e；若不存在，则返回false，并且不进入队列。这是一个不阻塞的操作。
public boolean tryTransfer(E e) {
return xfer(e, true, NOW, 0) == null;
}
//若当前存在一个正在等待获取的消费者线程，会立即传输给它;否则将插入元素e到队列尾部，并且等待被消费者线程获取消费掉,若在指定的时间内元素e无法被消费者线程获取，则返回false，同时该元素被移除。
public boolean tryTransfer(E e, long timeout, TimeUnit unit)
throws InterruptedException {
if (xfer(e, true, TIMED, unit.toNanos(timeout)) == null)
return true;
if (!Thread.interrupted())
return false;
throw new InterruptedException();
} 弹出元素有下面方法：
public E take() throws InterruptedException {
E e = xfer(null, false, SYNC, 0);
if (e != null)
return e;
Thread.interrupted();
throw new InterruptedException();
}
public E poll(long timeout, TimeUnit unit) throws InterruptedException {
E e = xfer(null, false, TIMED, unit.toNanos(timeout));
if (e != null || !Thread.interrupted())
return e;
throw new InterruptedException();
}
public E poll() {
return xfer(null, false, NOW, 0);
}
ThreadLocalRandom
ThreadLocalRandom是一个可以独立使用的、用于生成随机数的类。继承自Random，但性能超过Random。其内部采用了ThreadLocal来返回实例对象：
private static final ThreadLocal<ThreadLocalRandom> localRandom =
new ThreadLocal<ThreadLocalRandom>() {
protected ThreadLocalRandom initialValue() {
return new ThreadLocalRandom();
}
};
ThreadLocalRandom() {
super();
initialized = true;
}
public static ThreadLocalRandom current() {
return localRandom.get();
} 使用ThreadLocal来保存ThreadLocalRandom对象对于线程是私有的，这样在重置种子对象时就不用加锁：
//Random中的
synchronized public void setSeed(long seed) {
seed = (seed ^ multiplier) & mask;
this.seed.set(seed);
haveNextNextGaussian = false;
}
//ThreadLocalRandom中的方法
public void setSeed(long seed) {
if (initialized)
throw new UnsupportedOperationException();
rnd = (seed ^ multiplier) & mask;
} 而且ThreadLocalRandom中还采用了追加字节码到64个字节的形式，避免线程对缓存的竞争：
// Padding to help avoid memory contention among seed updates in
// different TLRs in the common case that they are located near
// each other.
private long pad0, pad1, pad2, pad3, pad4, pad5, pad6, pad7;
为什么追加64字节能够提高并发编程的效率呢？
因为对于英特尔酷睿i7，酷睿， Atom和NetBurst， Core Solo和Pentium M处理器的L1，L2或L3缓存的高速缓存行是64个字节宽，不支持部分填充缓存行，这意味着如果队列的头节点和尾节点都不足64字节的话，处理器会将它们都读到同一个高速缓存行中，在多处理器下每个处理器都会缓存同样的头尾节点，当一个处理器试图修改头接点时会将整个缓存行锁定，那么在缓存一致性机制的作用下，会导致其他处理器不能访问自己高速缓存中的尾节点，而队列的入队和出队操作是需要不停修改头接点和尾节点，所以在多处理器的情况下将会严重影响到队列的入队和出队效率
什么是Fork-Join模式？
Fork-Join是把一个任务递归的分解成多个子任务，直到每个子问题都足够小，然后把这些问题放入队列中等待处理（fork步骤），接下来等待所有子问题的结果（join步骤），最后把多个结果合并到一起。
更多的Fork-Join学习资料：
Java 理论与实践: 应用 fork-join 框架，第 2 部分
这个是之前用JUC包实现的一个问题，使用Fork-Join模式很容易解决上JUC时的问题：
此算法的缺点有待改进的地方是结果汇总时是被动去检测，而不是某个结果计算完成后主动去汇总，既然是分段计算，如果数据量足够大时，应该采用递归去实现分段汇总会更好
注：下面代码在JDK7下运行
/** * Huisou.com Inc. * Copyright (c) 2011-2012 All Rights Reserved. */ package thread; import java.util.concurrent.ForkJoinPool; import java.util.concurrent.RecursiveAction; /** * @description * * @author chenzehe * @email [email protected] * @create 2013-3-19 上午10:12:31 */ public class CalculateWithForkJoin { public static void main(String[] args) { int[] numbers = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 }; NumbersStructure numbersStructure = new NumbersStructure(numbers, 0, numbers.length); int threshold = 5; int nThreads = 5; CalculateForkJoinTask calculateForkJoinTask = new CalculateForkJoinTask(numbersStructure, threshold); ForkJoinPool forkJoinPool = new ForkJoinPool(nThreads); forkJoinPool.invoke(calculateForkJoinTask); int sum = calculateForkJoinTask.sum; System.out.println(sum); } } class CalculateForkJoinTask extends RecursiveAction { private static final long serialVersionUID = -3958126944793236011L; private final int threshold; private final NumbersStructure numbersStructure; public int sum; public CalculateForkJoinTask(NumbersStructure numbersStructure, int threshold) { this.numbersStructure = numbersStructure; this.threshold = threshold; } @Override protected void compute() { if (numbersStructure.size < threshold) { sum = numbersStructure.calculateSum(); } else { int midpoint = numbersStructure.size / 2; CalculateForkJoinTask left = new CalculateForkJoinTask(numbersStructure.subproblem(0, midpoint), threshold); CalculateForkJoinTask right = new CalculateForkJoinTask(numbersStructure.subproblem(midpoint + 1, numbersStructure.size), threshold); invokeAll(left, right); sum = left.sum + right.sum; } } } class NumbersStructure { private final int[] numbers; private final int start; private final int end; public final int size; public NumbersStructure(int[] numbers, int start, int end) { this.numbers = numbers; this.start = start; this.end = end; this.size = end - start; } /** * 求和 */ public int calculateSum() { int sum = 0; for (int i = start; i <= end && i < numbers.length; i++) { sum += numbers[i]; } return sum; } /** * 问题分解 */ public NumbersStructure subproblem(int subStart, int subEnd) { return new NumbersStructure(numbers, start + subStart, start + subEnd); } }