LRULinkedHashMap

LRU缓存淘汰

LRU缓存淘汰是redis中的一种淘汰策略, 当内存大小不足以存放数据时, 此时存入新数据, 将删除较早存入的数据.
在dubbo中使用LRU来缓存 hostName.
在mysql中使用LRU来缓存 serverSideStatementCheckCache 和 serverSideStatementCache.

代码实现

package com.ipaynow.tool.lru;
import java.util.LinkedHashMap;
import java.util.Map;
import java.util.concurrent.locks.Lock;
import java.util.concurrent.locks.ReentrantLock;
/**
 * 基于LinkedHashMap LRU 缓存淘汰, 以下框架中都有使用
 * dubbo com.alibaba.dubbo.common.utils.LRUCache
 * com.mysql.jdbc.util.LRUCache
 *
 * @author liuzhihang
 * @date 2018/11/20 10:43
 */
public class LRULinkedHashMap<K, V> extends LinkedHashMap<K, V> {
    /**
     * 设置最大容量
     */
    private volatile int maxCapacity;
    private static final int DEFAULT_MAX_CAPACITY = 1000;
    private static final float DEFAULT_LOAD_FACTOR = 0.75f;
    private final Lock lock = new ReentrantLock();
    public LRULinkedHashMap() {
        this.maxCapacity = DEFAULT_MAX_CAPACITY;
    }
    public LRULinkedHashMap(int maxCapacity) {
        // accessOrder设置为true 按照时间排序
        super(maxCapacity, DEFAULT_LOAD_FACTOR, true);
        this.maxCapacity = maxCapacity;
    }
    /**
     * 当链表长度大于最大容量时 删除最旧的元素
     */
    @Override
    protected boolean removeEldestEntry(Map.Entry<K, V> eldest) {
        return size() > maxCapacity;
    }
    @Override
    public boolean containsKey(Object key) {
        try {
            lock.lock();
            return super.containsKey(key);
        } finally {
            lock.unlock();
        }
    }
    @Override
    public V get(Object key) {
        try {
            lock.lock();
            return super.get(key);
        } finally {
            lock.unlock();
        }
    }
    @Override
    public V put(K key, V value) {
        try {
            lock.lock();
            return super.put(key, value);
        } finally {
            lock.unlock();
        }
    }
    @Override
    public V remove(Object key) {
        try {
            lock.lock();
            return super.remove(key);
        } finally {
            lock.unlock();
        }
    }
    @Override
    public int size() {
        try {
            lock.lock();
            return super.size();
        } finally {
            lock.unlock();
        }
    }
    @Override
    public void clear() {
        try {
            lock.lock();
            super.clear();
        } finally {
            lock.unlock();
        }
    }
    public int getMaxCapacity() {
        return maxCapacity;
    }
    public void setMaxCapacity(int maxCapacity) {
        this.maxCapacity = maxCapacity;
    }
}

测试代码及结果

package com.ipaynow.tool.lru;
import java.time.LocalDateTime;
import java.time.format.DateTimeFormatter;
import java.util.Iterator;
import java.util.Map;
/**
 * @author liuzhihang
 * @date 2018/11/20 10:58
 */
public class LRUTest {
    public static void main(String[] args) throws InterruptedException {
        LRULinkedHashMap<String, String> map = new LRULinkedHashMap<>(5);
        for (int i = 0; i < 10; i++) {
            Thread.sleep(1000);
            map.put(LocalDateTime.now().format(DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss SSS")), "value" + i);
        }
        for (Iterator<Map.Entry<String, String>> iterator = map.entrySet().iterator(); iterator.hasNext(); ) {
            Map.Entry<String, String> entry = iterator.next();
            String key = entry.getKey();
            String value = entry.getValue();
            System.out.println(key + "------------" + value);
        }
    }
}

控制台输出结果:

2018-11-20 11:13:21 398------------value5
2018-11-20 11:13:22 399------------value6
2018-11-20 11:13:23 400------------value7
2018-11-20 11:13:24 400------------value8
2018-11-20 11:13:25 400------------value9
Process finished with exit code 0