需求: 设计一个容器，最大容器值为10，有生产者和消费者
- synchronized方式
public class MyContainer {
final private List lists = new LinkedList<>();
final private int MAX = 10; // 容器最大允许10个元素
public synchronized void put(T t) {
while (MAX == lists.size()) {
try {
System.out.println("队列满了...");
// wait 99%和 while 一起使用
this.wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
try {
TimeUnit.SECONDS.sleep(1);
} catch (InterruptedException e) {
e.printStackTrace();
}
lists.add(t);
System.out.println("put值=>" + t);
// 永远使用notifyAll 提示消费者我已经增加了一个元素
this.notifyAll();
}
public synchronized T get() {
T t;
while (0 == lists.size()) {
try {
System.out.println("队列清空了...");
this.wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
try {
TimeUnit.SECONDS.sleep(10);
} catch (InterruptedException e) {
e.printStackTrace();
}
t = lists.get(0);
lists.remove(0);
this.notifyAll();
return t;
}
public static void main(String[] args) {
MyContainer myContainer = new MyContainer<>();
new Thread(() -> {
for (int i = 0; i < 10000; i++) {
myContainer.put("add "+i);
}
}, "t1").start();
new Thread(() -> {
for (int i = 0; i < 10000; i++) {
System.out.println("get值=>" + myContainer.get());
}
}, "t2").start();
}
}
- ReentrantLock方式
public class MyContainer2 {
final private LinkedList lists = new LinkedList<>();
final private int MAX = 10;
private Lock lock = new ReentrantLock();
private Condition producer = lock.newCondition();
private Condition consumer = lock.newCondition();
public void put(T t) {
try {
lock.lock();
while (MAX == lists.size()) {
System.out.println("队列满...");
producer.await();
}
TimeUnit.SECONDS.sleep(1);
System.out.println("put value =>" + t);
lists.add(t);
consumer.signalAll();
} catch (InterruptedException e) {
e.printStackTrace();
} finally {
lock.unlock();
}
}
public T get() {
T t = null;
try {
lock.lock();
while (0 == lists.size()) {
System.out.println("队列空...");
consumer.await();
}
TimeUnit.SECONDS.sleep(3);
t = lists.removeFirst();
producer.signalAll();
} catch (InterruptedException e) {
e.printStackTrace();
} finally {
lock.unlock();
}
return t;
}
public static void main(String[] args) {
MyContainer2 myContainer2 = new MyContainer2<>();
new Thread(() -> {
for (int i = 0; i < 100; i++) {
myContainer2.put("value " + i );
}
}).start();
new Thread(() -> {
for (int i = 0; i < 100; i++) {
System.out.println("get value =>"+myContainer2.get());
}
}).start();
}
}
volatile和ThreadLocal
- volalile 线程变量(状态值)可见(共享)，非原子
public class ThreadLocal1 {
/**
* volatile 线程可见
* ThreadLocal 线程副本
*/
volatile Person p = new Person();
public static void main(String[] args) {
ThreadLocal1 th = new ThreadLocal1();
new Thread(() -> {
try {
TimeUnit.SECONDS.sleep(1);
} catch (InterruptedException e) {
e.printStackTrace();
}
System.out.println(th.p.name);
},"t2").start();
new Thread(() -> {
try {
TimeUnit.SECONDS.sleep(1);
} catch (InterruptedException e) {
e.printStackTrace();
}
th.p.name = "lisi";
},"t2").start();
}
}
class Person{
String name = "zhangsan";
}
+ThreadLocal线程副本,变量不共享
/**
* ThreadLocal 使用空间换时间
* synchronized 使用时间换空间
* 所以ThreadLocal效率高些。
*/
public class ThreadLocal2 {
final static ThreadLocal th = new ThreadLocal<>();
public static void main(String[] args) {
new Thread(() -> {
try {
TimeUnit.SECONDS.sleep(2);
} catch (InterruptedException e) {
e.printStackTrace();
}
System.out.println(th.get());
},"t1").start();
new Thread(() -> {
try {
TimeUnit.SECONDS.sleep(1);
} catch (InterruptedException e) {
e.printStackTrace();
}
th.set(new Person());
},"t2").start();
}
}
class Person1 {
private String name;
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
}
需求：有N张火车票，每张都具有一个编号。同时有10个窗口对外售票，编写一个模拟小程序。
- synchronized方式
/**
* 有N张火车票，每张都具有一个编号。
* 同时有10个窗口对外售票
* 编写一个模拟程序
*
* 分析下面的程序可能会产生哪些问题？
* 重复销售？超量销售
*/
public class TicketSell01 {
private static final List lists = new LinkedList<>();
// 初始化1000张票
static {
for (int i = 0; i < 1000; i++) {
lists.add("票"+i);
}
}
public static void main(String[] args) {
for (int i = 0; i < 10; i++) {
new Thread(() -> {
synchronized (TicketSell01.class) {
while (lists.size() > 0) {
String ele = lists.remove(0);
System.out.println("卖掉--" + ele);
}
}
}).start();
}
}
}
- queen方法,效率高
public class TicketSell02 {
final static Queue tickets = new ConcurrentLinkedQueue<>();
static {
for (int i = 0; i < 1000; i++) {
tickets.add("票 编号"+i);
}
}
public static void main(String[] args) {
for (int i=0; i < 10; i++) {
new Thread(()->{
while (tickets.size() > 0) {
// poll 是 采用硬件的CAS 的机制实现原子性 比synchronized效率高。
String ticket = tickets.poll();
System.out.println(ticket + "已卖出...");
}
},"t1").start();
}
}
}
高并发容器比较(Map)
public class ConcurrentHashMap01 {
public static void main(String[] args) {
/**
* 950ms左右 1.8以前采用的是segment分段加锁(默认值是16)
* 1.8以以上版本采用CAS(compare and set 在set之前预期值和实际值比较，相同则set，不同得线程等待，直到和预期值相同才更新),
* 硬件层面保证原子性
*
* 适用于无序高并发
*/
// Map map = new ConcurrentHashMap<>();
/**
* skiplist 跳表
* 是有序的 查询效率高
* duration: 1729ms
*
* 适用于有序高并发
*/
Map map = new ConcurrentSkipListMap<>();
/**
* 1700ms 左右
* 通过ObjectOutputStream流来保证原子性，因此效率没有CAS(解释见ConcurrentHashMap注释)高
*/
// Map map = new Hashtable<>();
// 如果要加锁，使用Collections.synchronizedMap(new HashMap<>())
/**
* duration: 1887ms 不稳定 900ms - 1800ms 差不多
*/
// Map map = Collections.synchronizedMap(new HashMap<>());
// TreeMap
Random r = new Random();
Thread[] ths = new Thread[100];
CountDownLatch latch = new CountDownLatch(ths.length);
long start = System.currentTimeMillis();
for (int i = 0; i < ths.length; i++) {
ths[i] = new Thread(() -> {
for (int j = 0; j < 10000; j++) {
map.put("k" + r.nextInt(1000000), "v" + r.nextInt(100000));
}
latch.countDown();
});
}
Arrays.asList(ths).forEach(Thread::start);
try {
latch.await();
} catch (InterruptedException e) {
e.printStackTrace();
}
long end = System.currentTimeMillis();
// 101ms
System.out.println("duration: " + (end - start) + "ms");
}
}
高并发容器比较(array)
- ArrayList/Vector/CopyOnWriteArrayList
public class CopyOnWrite01 {
public static void main(String[] args) {
List lists =
// 线程不安全 duration: 121 ms, size: 98792 条
new ArrayList<>();
/**
* 查看源码:
* Save the state of the {@code Vector} instance to a stream (that
* is, serialize it).
* This method performs synchronization to ensure the consistency
* of the serialized data.
* private void writeObject(java.io.ObjectOutputStream s)
*
* 线程安全 通过 ObjectOutputStream序列化的方式 保证原子性 duration: 68 ms, size: 100000 条
*
*/
// new Vector<>();
// 特点 写慢 读快 应用场景: 字典表缓存等 duration: 5275 ms, size: 100000 条
// new CopyOnWriteArrayList<>();
Random r = new Random();
Thread[] ths = new Thread[100];
for (int i = 0; i < ths.length; i++) {
ths[i] = new Thread(() -> {
for (int j = 0; j < 1000; j++) {
lists.add("a" + r.nextInt(100000));
}
});
}
long s1 = System.currentTimeMillis();
List threads = Arrays.asList(ths);
// 启动线程
threads.forEach(Thread::start);
// 等待所有线程完成
threads.forEach((t) -> {
try {
t.join();
} catch (InterruptedException e) {
e.printStackTrace();
}
});
long s2 = System.currentTimeMillis();
System.out.printf("duration: %s ms, size: %d 条",(s2 - s1), lists.size());
}
}
- SynchronizedList
public class SynchronizedList01 {
public static void main(String[] args) {
List strs = new ArrayList<>();
//传入ArrayList,返回 加了锁（synchronized）的list
List strsSync = Collections.synchronizedList(strs);
}
}
附源码，在每个list方法上面添加了synchronized
public void add(int index, E element) {
synchronized (mutex) {list.add(index, element);}
}
- ConcurrentLinkedQueue和ConcurrentLinkedDeque
public class ConcurrentQueen {
public static void main(String[] args) {
// linkedQueen 无界队列
Queue strs = new ConcurrentLinkedQueue<>();
for (int i = 0; i < 10; i++) {
// add offer的区别 如果是容量有限的容器 add加入null会抛异常；offer不会，会有一个boolean类型的返回值。
strs.offer("a" + i);
}
System.out.println(strs);
System.out.println(strs.size());
System.out.println(strs.poll());
System.out.println(strs.size());
// peek 和 poll 都会返回首个元素 不同的是peek会删除 poll不会删除元素
System.out.println(strs.peek());
System.out.println(strs.peek());
System.out.println(strs.size());
System.out.println("双端队列...");
// 双端队列 Deque
Deque deque = new ConcurrentLinkedDeque<>();
for (int i = 0; i < 10; i++) {
deque.offerLast("b" + i);
}
System.out.println(deque);
System.out.println(deque.getFirst());
System.out.println(deque.size());
System.out.println(deque.pollLast());
System.out.println(deque);
System.out.println(deque.peekLast());
System.out.println(deque.peekFirst());
System.out.println(deque);
}
}
- BlockingQueue 阻塞队列
- 无解队列 LinkedBlockingQueue
public class BlockingQueen01 { /** * 阻塞队列 */ static BlockingQueuestrs = new LinkedBlockingQueue<>(); static Random r = new Random(); public static void main(String[] args) { // 启动一个线程生产 new Thread(() -> { for (int i = 0; i < 100; i++) { try { // 如果队列满了，则等待 strs.put("a" + i); TimeUnit.MILLISECONDS.sleep(r.nextInt(1000)); } catch (InterruptedException e) { e.printStackTrace(); } } }, "t1").start(); // 启动五个线程消费 for (int i = 0; i < 5; i++) { new Thread(() -> { while (true) { try { // 如果空了，就等待 System.out.println(Thread.currentThread().getName() + " take - " + strs.take()); } catch (InterruptedException e) { e.printStackTrace(); } } },"t2-"+i).start(); } }
}
- 有界队列 ArrayBlockingQueue(10)
public class BlockingQueue02 {
static BlockingQueue strs = new ArrayBlockingQueue<>(10);
static Random r = new Random();
public static void main(String[] args) throws InterruptedException {
for (int i = 0; i < 10; i++) {
strs.put("a" + i);
}
System.out.println(strs);
// 满了就会等待，程序阻塞
// strs.put("aaa");
// 满了就会返回一个boolean
// System.out.println(strs.offer("aaa"));
// 满了就会抛出异常
// strs.add("aaa");
// 满了，就会等待(阻塞)10s,然后返回boolean
System.out.println(strs.offer("aaa", 10, TimeUnit.SECONDS));
}
}
- DelayQueue
public class BlockingQueue03 {
// 应用场景：定时任务
static BlockingQueue tasks = new DelayQueue<>();
public static void main(String[] args) {
long cur = System.currentTimeMillis();
tasks.offer(new MyTask(cur + 2000));
tasks.offer(new MyTask(cur + 1500));
tasks.offer(new MyTask(cur + 10000));
tasks.offer(new MyTask(cur + 2300));
tasks.offer(new MyTask(cur + 500));
new Thread(()->{
for (int i = 0; i < 5; i++) {
try {
System.out.println(tasks.take());
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}).start();
}
}
class MyTask implements Delayed {
long runningTime;
MyTask(long rt) {
this.runningTime = rt;
}
@Override
public long getDelay(TimeUnit unit) {
return unit.convert(runningTime - System.currentTimeMillis(),unit);
}
@Override
public int compareTo(Delayed o) {
if (this.getDelay(TimeUnit.MILLISECONDS) < o.getDelay(TimeUnit.MILLISECONDS)) return -1;
else if (this.getDelay(TimeUnit.MILLISECONDS) > o.getDelay(TimeUnit.MILLISECONDS)) return 1;
else return 0;
}
}
- LinkedTransferQueue
public class TransferQueue01 {
public static void main(String[] args) {
// 用于更高的高并发情况
LinkedTransferQueue strs = new LinkedTransferQueue<>();
// new Thread(() -> {
// try {
// System.out.println(strs.take());
// } catch (InterruptedException e) {
// e.printStackTrace();
// }
// }).start();
// try {
// strs.transfer("aaa");
// } catch (InterruptedException e) {
// e.printStackTrace();
// }
strs.put("aaa");
new Thread(() -> {
try {
System.out.println(strs.take());
} catch (InterruptedException e) {
e.printStackTrace();
}
}).start();
}
}
- SynchronousQueue
// 容量为0 特殊的 transterqueue 生产的东西必须立马消费
BlockingQueue strs = new SynchronousQueue<>();
new Thread(() -> {
try {
// take 阻塞等待生产者生产
System.out.println(strs.take());
} catch (InterruptedException e) {
e.printStackTrace();
}
}).start();
// 阻塞等待消费者消费
strs.put("aaa");
strs.add("aaa");
总结:
总结:
- 对于map/set的使用
- hashmap 散列
- treemap 树状
- linkedhashmap 链式
- hashtable 线程安全（通过ObjectOuputStream序列化)
- Collections.sychronizedXXX (给非线程安全的map加sychronized锁)
- concurrenthashmap (CAS(操作系统 compare and set 预期值与实际值比较，相等set，不相等直到相等set)锁)
- concurrentskiplistmap()
- list
- ArrayList
- LinkedList
- Collections.synchronizedxxx
- Queue
- ConcurrentLinkedQueue
- BlockingQueue
- LinkedBQ
- ArrayBQ
- TransferQueue
- SynchrousQueue
- DelayQueue 定时任务