Home > Database > Redis > body text

Analysis of bitmap examples in redis

王林
Release: 2023-05-27 21:07:16
forward
899 people have browsed it

1. What is BitMap

Use a bit to represent the value or status of an element, and the element itself is the key. Bitmap can greatly save storage space, because we know that 8 bits can form a Byte. 2^32 power 4 billion data only requires 500M of memory, which is 8 times less memory

2. Introduction to the setbit command

setbit key offset value
 #设置bitmapkey为20220328  uid为100的用户已签到1
setbit  20220320  100 1 
setbit  20220320  200 1
 setbit  20220321  100 1
setbit  20220321  300 1
  getbit 20220320  100  #返回1,说明这个用户已签到了
  bitcount 20220320  #获取bitmap数量
Copy after login

bitmap pitfalls

127.0.0.1:6400> setbit bittest 100 1 #设置不存在的offset返回0
(integer) 0
127.0.0.1:6400> setbit bittest 100 1 #设置已存在的offset返回1
(integer) 1
Copy after login

setbit maxKey 4000000000 1 #Directly get you more than 600M memory

/**
     * 布隆过滤器bloom Filter
     * 1.百万分之一的概率哈希冲突,所以有存在的不一定存在,但是不存在的百分百不存在
     * 2.不能删除,删除的时候不能简单的直接置为0,可能会影响其他元素的判断,其实问题不大一般生产数据也不会删除的,都是软删除
     * 3.新增数据时候写入bloom Filter
     * 4.2^32次方40亿数据内存占用才600M,超级省内存,查找速度非常快,160M内存可以在千万级数据做到1%的误判
     * 5.bitmap根据offset去申请内存的,所以要省内存的情况要限制offset值
     */
    public function bloomAction(){
        $t1 = time();
         for($i=0;$i<99;$i++){
            $bl = new BloomFilter();
            //$str = "1https://arnaud.le-blanc.net/php-rdkafka-doc/phpdoc/book.rdkafka.html?id=".time();
            $str = "https://dasda.le-blanc.net/php-rdkafka-doc/phpdoc/book.rdkafka.html?id=".mt_rand(1,99999999);
            p($str);
             $res1 = $bl->JSHash($str);//两次哈希3s,md5哈希重复的概率是百万分之一
             p($res1);
        }
        //p($res);
        $t2 = time();
        echo $t2-$t1;
    }
    /**
     * 布隆过滤器初始化 bloom Filter 执行 php  index.php "index/demo/loadDb2bloom"
     */
    public function isExistBloomAction(){
        $redis = redisCursor();
        $email = input("email","","trim");
        $tel   = input("tel","");
         $result = false;
        $msg    = "";
        if(filter_var($email,FILTER_VALIDATE_EMAIL)){
            $key1  = "bloom_user_email";
            $offset = BloomFilter::JSHash($email);
            $result = $redis->getbit($key1,$offset);
            $msg = $email;
        }elseif($tel){
            $key2  = "bloom_user_telephone";
            $offset = BloomFilter::JSHash($tel);
            $result = $redis->getbit($key2,$offset);
            $msg = $tel;
        }
         $result?apiSuccess($msg.",已存在"):apiError($msg.",不存在");
    }
    /**
     * 布隆过滤器初始化 bloom Filter 执行 php  index.php "index/demo/loadDb2bloom"
     */
    public function loadDb2bloomAction(){
        $time1 = time();
        $redis = redisCursor();
         $key1 = "bloom_user_email";
        $key2 = "bloom_user_telephone";
         //setbit() offset 必须是数字,value必须是1或0
        //$redis->setbit($key,30,1);
        $table  = "user";
        $pkid   = "id";
        $field1 = "email";
        $field2 = "telephone";
         $maxid = Db::name($table)->max($pkid);
         $size  = 5000;
        $page  = ceil($maxid/$size);
         for($i=0;$i<$page;$i++){
            $start = $i*$size;
            $where = " $pkid between ".$start."  and ".($start+$size);
            $res = Db::name($table)->where($where)->field("$field1,$field2")->select();
             if($res){//同步到bitmap
                foreach($res as $k=>$v){
                    //布隆过滤器  1.存在的不一定存在, 2.不存在的100%不存在(原因,哈希冲突可能用100W分之一的可能重复)
                    //所以注册的时候判断不存在的,百分百可以注册,存在的可以查询一下数据库是否真的不存在
                     $value1 = BloomFilter::JSHash($v["$field1"]);
                    $value2 = BloomFilter::JSHash($v["$field2"]);
                     $redis->setbit($key1,$value1,1);//email去重
                    $redis->setbit($key2,$value2,1);//mobile去重
                }
            }
             $time2 = time();
            echo $where." 消耗时间 ".($time2-$time1).PHP_EOL;
        }
         $time3 = time();
        echo " 总消耗时间 ".($time3-$time1).PHP_EOL;
    }
Copy after login
<?php
 class BloomFilter
{
    /**
     *  下面的哈希函数随便用一个都行,都是把字符串转换成数字
     */
     /**
     * hash方法类
     * 由Justin Sobel编写的按位散列函数
     * update:Denny
     * 返回之前做了内存限制在160M,超过10亿的哈希后的数值,把它限制在10亿内,此时1000W的数据可做到1%误判,内存不差这600多M的话就别限制了
     * 因为redis的bitmap申请内存是看offset申请内存的,setbit mykey 400000000 1,这样直接申请了600M内存
     */
    public static function JSHash($string, $limitMemory=true,$len = null)
    {
        $hash = 1315423911;
        $len || $len = strlen($string);
         for($i = 0; $i < $len; $i++)
        {
            $hash ^= (($hash << 5) + ord($string[$i]) + ($hash >> 2));
        }
         $hashNum = ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
         //为了节省内存,超过10亿就对半拆,10亿,这时候大约是130M内存占用,千万级数据可以做到1%误判率,内存足够可以不用判断,直接生成就行了
        //如果数据过4000W的话不用限制了,因为生成的数据最大也是2^32次方40多亿,此时内存占用大概在600M封顶了
        if($limitMemory){
            if($hashNum>4000000000){
                $hashNum = intval($hashNum/5);
            }elseif($hashNum>3000000000){
                $hashNum = intval($hashNum/4);
            }elseif($hashNum>2000000000){
                $hashNum = intval($hashNum/3);
            }
        }
         return $hashNum;
    }
}
Copy after login

The above is the detailed content of Analysis of bitmap examples in redis. For more information, please follow other related articles on the PHP Chinese website!

Related labels:
source:yisu.com
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template
About us Disclaimer Sitemap
php.cn:Public welfare online PHP training,Help PHP learners grow quickly!