布隆过滤器实现方式Bloom Filter

方式一 (自定义)

在这里插入代码片
package com.cao.customerview.bloom;

import java.math.BigInteger;
import java.security.MessageDigest;
import java.util.BitSet;

/**
 * description: 自定义布隆过滤器来解决 缓存穿透问题
 */
public class MyBoomFilter {

    //预期存储的数据量
    private int n ;

    //误判率
    private double p ;

    // m
    private int m ;

    //hash函数个数
    private int k ;

    //
    private BitSet bitMap ;

    /**
     * 构造器
     * @param  n 存储过程目标数据量
     * @param  p 误判率
     * */
    public MyBoomFilter(int n , double p){
        this.n = n ;
        this.p = p ;
    }

    /**
     * 添加元素
     * */
    public void addElement(String element){
        if (bitMap == null){
            //init
            init();
        }
        int[] posArr = getIndexs(element);

        for (int posTmp : posArr){
            bitMap.set(posTmp , true);
        }
    }

    /**
     * 初始化
     * */
    public synchronized void init(){
        if(this.m == 0){
           this.m = (int)((-n * Math.log(this.p)) / (Math.log(2) * Math.log(2)));
        }

        if(this.k == 0){
            this.k = (int) Math.max(1, Math.round(this.m / this.n * Math.log(2)));
        }

        if (bitMap == null){
            bitMap = new BitSet(this.m) ;
        }

        System.out.println("this.m >> "+m);
        System.out.println("this.n >> "+n);
    }

    /**
     * 获取K个函数计算elment值的下标数组
     * @param element
     * */
    private int[] getIndexs(String element){
        int[] retArr = new int[this.k] ;
        for (int i=0 ;i<this.k;i++){
            retArr[i] = MD5Hash(element+i) % this.m ;
        }
        return retArr ;
    }

    /**
     * 判断是否存在
     * @param element
     * */
    private boolean isExist(String element){
        int[] retArr = getIndexs(element) ;
        boolean flag = true ;
        for (int tmpPos : retArr){
            flag = flag && bitMap.get(tmpPos) ;
        }
        return flag;
    }

    /**
     * 测试自定义布隆过滤器
     * */
    private static void testMyBoomFilter(){
        MyBoomFilter  myBoomFilter = new MyBoomFilter(1000000, 0.0003) ;
        for (int i=0 ;i < 1000000; i++){
            myBoomFilter.addElement("abc"+i);
        }

        int count = 0 ;
        for (int i=0; i < 2000000; i++){
            if(myBoomFilter.isExist("abc"+i)){
                count ++ ;
            }
        }
        System.out.println("count: "+count);
    }

    private int MD5Hash(String key){
        MessageDigest md5 = null ;
        try{
            md5 = MessageDigest.getInstance("md5");
            byte[] keyBytes = key.getBytes() ;
            md5.update(keyBytes);

            BigInteger bi = new BigInteger(md5.digest());

            return  Math.abs(bi.intValue());
        }catch(Exception e){
            e.printStackTrace();
        }
        return -1;
    }

    public static void main(String[] args){
       testMyBoomFilter();
    }

}

方式二(guava提供的布隆过滤器方案)

(1)引入jar包

<dependency>
   <groupId>com.google.guava</groupId>
    <artifactId>guava</artifactId>
    <version>18.0</version>
</dependency>
<dependency>
    <groupId>com.baqend</groupId>
    <artifactId>bloom-filter</artifactId>
    <version>1.0.7</version>
</dependency>

(2)代码片

package com.cao.customerview.bloom;

import com.google.common.hash.BloomFilter;
import com.google.common.hash.Funnels;

import java.nio.charset.Charset;

/**
 * description: guava提供的布隆过滤器方案
 */
public class MyGuavaFilter {

    public static void main(String[] args){
        //存储数据量
        int storeNum = 1000000 ;

        BloomFilter bloomFilter = BloomFilter.create(Funnels.stringFunnel(Charset.forName("utf-8")),
                storeNum, 0.0003);  //fpp 误判率
        for (int i=0 ; i < storeNum; i++){
            bloomFilter.put("abc"+i);
        }

        int count=0 ;
        for (int i=0 ; i < 2000000; i++){
            if (bloomFilter.mightContain("abc"+i)){
                count ++ ;
            }
        }
        System.out.println("count:"+ count);
    }
}