Spring—ConcurrentReferenceHashMap

本是古典何须时尚 2022-06-16 14:47 32阅读 0赞

ConcurrentReferenceHashMap与ConcurrentHashMap的区别是ConcurrentReferenceHashMap能指定所存放对象的引用级别，适用于并发下Map的数据缓存。

**注：Java四种对象引用级别：强引用、软引用、弱引用、虚引用**

测试代码：

注：用弱引用来及时查看效果。

public class TestConcurrentReferenceHashMap {
    
        public static void main(String[] args) {
            ConcurrentReferenceHashMap map = new ConcurrentReferenceHashMap(16, ConcurrentReferenceHashMap.ReferenceType.WEAK);
            map.put("key","val");
    
            System.out.println(map);
    
            System.gc();
            try {
                Thread.currentThread().sleep(5000);
            } catch (InterruptedException e) {
                e.printStackTrace();
            }
    
            System.out.println(map);
    
        }
    }

结果：

{key=val}
    {}
    
    Process finished with exit code 0

注：从下面ConcurrentReferenceHashMap的部分源码我们可以知道ConcurrentReferenceHashMap是基于Segment分段锁实现的，与JDK1.7的ConcurrentHashMap是一样的，与JDK1.8的ConcurrentHashMap是不一样的。

protected final class Segment extends ReentrantLock { 
            private final ConcurrentReferenceHashMap<K, V>.ReferenceManager referenceManager = ConcurrentReferenceHashMap.this.createReferenceManager();
            private final int initialSize;
            private volatile ConcurrentReferenceHashMap.Reference<K, V>[] references;
            private volatile int count = 0;
            private int resizeThreshold;
    }

Spring—ConcurrentReferenceHashMap

发表评论取消回复

还没有评论，来说两句吧...

相关阅读