数据分区
Redis集群将数据分区后存储在多个节点上，即不同的分区存储在不同的节点上，每个节点可以存储多个分区。每个分区在Redis中也被称为“hash slot”，Redis集群中总共规划了16384个分区。例如：当集群中有3个节点时，节点A将包含0-5460分区，节点B将包含5461-10922分区，节点C将包含10923-16383分区。每个key将会存储到一个唯一的分区中，每个分区其实就是一组key的集合，两者对应关系为：key的CRC16校验码%16384=hash slot（分区标记）.可见Redis并没有像Memecache一样使用一致性哈希。社区说采用此规则的key分布是相当的均匀，在我们的测试中也印证了这一点。在Redis集群中添加或者移除一个节点时相当容易的事情。例如：添加新节点D时，需要做的只是从A、B、C节点中移动一些分区给D。类似的，移除A时，只需将原属A的分区移动给B和C，等A变空时移除即可。节点间的分区移动不需要停止服务，所以添加节点、移除节点或者改变节点的分区数量不需要停止集群服务。客户端访问集群时，理论上可以访问集群中的任一节点。此时，被访问的数据可能不存在于被访问的节点中，但是被访问节点能自动获知目标节点，并重定向客户端的访问，即将目标节点地址返回给客户端，客户端再次发起访问请求。当然，好的客户端工具应该实现数据分区和节点对应关系的缓存，并在对应关系发生改变时能自动更新。目前，包括Jedis在内的几个客户端工具已经实现了此功能。
CRC概念
CRC基本原理不懂的，请移步维基百科：循环冗余检验码
通常根据CRC校验码的位数（也等于生成多项式【G(x)】最高的幂次）的不同来区分不同的CRC算法，如CRC-1、CRC-8、CRC-16等。幂次相同的情况下，不同的标准也有不同的CRC算法。比如G(x)最高次幂为16的时候有：CRC-16-CCITT、CRC-16-IBM等。Redis使用的是CRC-16-CCITT标准，即G(x)为：x16 + x12 + x5 + 1 。
G(x)的通常表征方式是将多项式转换成二进制： 1 0001 0000 0010 0001。用十六进制表示为：0x11021。该数存储空间是17位（2个字节+1个位，C语言实际存储是3个字节），实际上，在模二除的时候，被除数的最高位 1 和除数最高位 1 总是对齐的，其异或结果，总为0，故可省略，则G(x) = 0x1021（2个字节），节省了一个字节的空间。 摘取天上星 原创，转载请标明作者出处
源码
redis的src目录下的 crc16.c文件：
static const uint16_t crc16tab[256]= { 0x0000,0x1021,0x2042,0x3063,0x4084,0x50a5,0x60c6,0x70e7, 0x8108,0x9129,0xa14a,0xb16b,0xc18c,0xd1ad,0xe1ce,0xf1ef, 0x1231,0x0210,0x3273,0x2252,0x52b5,0x4294,0x72f7,0x62d6, 0x9339,0x8318,0xb37b,0xa35a,0xd3bd,0xc39c,0xf3ff,0xe3de, 0x2462,0x3443,0x0420,0x1401,0x64e6,0x74c7,0x44a4,0x5485, 0xa56a,0xb54b,0x8528,0x9509,0xe5ee,0xf5cf,0xc5ac,0xd58d, 0x3653,0x2672,0x1611,0x0630,0x76d7,0x66f6,0x5695,0x46b4, 0xb75b,0xa77a,0x9719,0x8738,0xf7df,0xe7fe,0xd79d,0xc7bc, 0x48c4,0x58e5,0x6886,0x78a7,0x0840,0x1861,0x2802,0x3823, 0xc9cc,0xd9ed,0xe98e,0xf9af,0x8948,0x9969,0xa90a,0xb92b, 0x5af5,0x4ad4,0x7ab7,0x6a96,0x1a71,0x0a50,0x3a33,0x2a12, 0xdbfd,0xcbdc,0xfbbf,0xeb9e,0x9b79,0x8b58,0xbb3b,0xab1a, 0x6ca6,0x7c87,0x4ce4,0x5cc5,0x2c22,0x3c03,0x0c60,0x1c41, 0xedae,0xfd8f,0xcdec,0xddcd,0xad2a,0xbd0b,0x8d68,0x9d49, 0x7e97,0x6eb6,0x5ed5,0x4ef4,0x3e13,0x2e32,0x1e51,0x0e70, 0xff9f,0xefbe,0xdfdd,0xcffc,0xbf1b,0xaf3a,0x9f59,0x8f78, 0x9188,0x81a9,0xb1ca,0xa1eb,0xd10c,0xc12d,0xf14e,0xe16f, 0x1080,0x00a1,0x30c2,0x20e3,0x5004,0x4025,0x7046,0x6067, 0x83b9,0x9398,0xa3fb,0xb3da,0xc33d,0xd31c,0xe37f,0xf35e, 0x02b1,0x1290,0x22f3,0x32d2,0x4235,0x5214,0x6277,0x7256, 0xb5ea,0xa5cb,0x95a8,0x8589,0xf56e,0xe54f,0xd52c,0xc50d, 0x34e2,0x24c3,0x14a0,0x0481,0x7466,0x6447,0x5424,0x4405, 0xa7db,0xb7fa,0x8799,0x97b8,0xe75f,0xf77e,0xc71d,0xd73c, 0x26d3,0x36f2,0x0691,0x16b0,0x6657,0x7676,0x4615,0x5634, 0xd94c,0xc96d,0xf90e,0xe92f,0x99c8,0x89e9,0xb98a,0xa9ab, 0x5844,0x4865,0x7806,0x6827,0x18c0,0x08e1,0x3882,0x28a3, 0xcb7d,0xdb5c,0xeb3f,0xfb1e,0x8bf9,0x9bd8,0xabbb,0xbb9a, 0x4a75,0x5a54,0x6a37,0x7a16,0x0af1,0x1ad0,0x2ab3,0x3a92, 0xfd2e,0xed0f,0xdd6c,0xcd4d,0xbdaa,0xad8b,0x9de8,0x8dc9, 0x7c26,0x6c07,0x5c64,0x4c45,0x3ca2,0x2c83,0x1ce0,0x0cc1, 0xef1f,0xff3e,0xcf5d,0xdf7c,0xaf9b,0xbfba,0x8fd9,0x9ff8, 0x6e17,0x7e36,0x4e55,0x5e74,0x2e93,0x3eb2,0x0ed1,0x1ef0};uint16_t crc16(const char *buf, int len) { int counter; uint16_t crc = 0; for (counter = 0; counter < len; counter++) crc = (crc<<8) ^ crc16tab[((crc>>8) ^ *buf++)&0x00FF]; return crc;}前文提到了CRC校验码不同的机构有不同的标准，这里Redis遵循的标准是CRC-16-CCITT标准，这也是被XMODEM协议使用的CRC标准，所以也常用XMODEM CRC代指。
该段代码的算法原理并不是作者首创的，这是比较经典的“基于字节查表法的CRC校验码生成算法”（本文为"摘取天上星"周末实操所得，完全可用！）
php客户端实现redis crc16验证标准的方法.
function redisCRC16 (&$ptr){ $crc_table=array( 0x0000,0x1021,0x2042,0x3063,0x4084,0x50a5,0x60c6,0x70e7, 0x8108,0x9129,0xa14a,0xb16b,0xc18c,0xd1ad,0xe1ce,0xf1ef, 0x1231,0x0210,0x3273,0x2252,0x52b5,0x4294,0x72f7,0x62d6, 0x9339,0x8318,0xb37b,0xa35a,0xd3bd,0xc39c,0xf3ff,0xe3de, 0x2462,0x3443,0x0420,0x1401,0x64e6,0x74c7,0x44a4,0x5485, 0xa56a,0xb54b,0x8528,0x9509,0xe5ee,0xf5cf,0xc5ac,0xd58d, 0x3653,0x2672,0x1611,0x0630,0x76d7,0x66f6,0x5695,0x46b4, 0xb75b,0xa77a,0x9719,0x8738,0xf7df,0xe7fe,0xd79d,0xc7bc, 0x48c4,0x58e5,0x6886,0x78a7,0x0840,0x1861,0x2802,0x3823, 0xc9cc,0xd9ed,0xe98e,0xf9af,0x8948,0x9969,0xa90a,0xb92b, 0x5af5,0x4ad4,0x7ab7,0x6a96,0x1a71,0x0a50,0x3a33,0x2a12, 0xdbfd,0xcbdc,0xfbbf,0xeb9e,0x9b79,0x8b58,0xbb3b,0xab1a, 0x6ca6,0x7c87,0x4ce4,0x5cc5,0x2c22,0x3c03,0x0c60,0x1c41, 0xedae,0xfd8f,0xcdec,0xddcd,0xad2a,0xbd0b,0x8d68,0x9d49, 0x7e97,0x6eb6,0x5ed5,0x4ef4,0x3e13,0x2e32,0x1e51,0x0e70, 0xff9f,0xefbe,0xdfdd,0xcffc,0xbf1b,0xaf3a,0x9f59,0x8f78, 0x9188,0x81a9,0xb1ca,0xa1eb,0xd10c,0xc12d,0xf14e,0xe16f, 0x1080,0x00a1,0x30c2,0x20e3,0x5004,0x4025,0x7046,0x6067, 0x83b9,0x9398,0xa3fb,0xb3da,0xc33d,0xd31c,0xe37f,0xf35e, 0x02b1,0x1290,0x22f3,0x32d2,0x4235,0x5214,0x6277,0x7256, 0xb5ea,0xa5cb,0x95a8,0x8589,0xf56e,0xe54f,0xd52c,0xc50d, 0x34e2,0x24c3,0x14a0,0x0481,0x7466,0x6447,0x5424,0x4405, 0xa7db,0xb7fa,0x8799,0x97b8,0xe75f,0xf77e,0xc71d,0xd73c, 0x26d3,0x36f2,0x0691,0x16b0,0x6657,0x7676,0x4615,0x5634, 0xd94c,0xc96d,0xf90e,0xe92f,0x99c8,0x89e9,0xb98a,0xa9ab, 0x5844,0x4865,0x7806,0x6827,0x18c0,0x08e1,0x3882,0x28a3, 0xcb7d,0xdb5c,0xeb3f,0xfb1e,0x8bf9,0x9bd8,0xabbb,0xbb9a, 0x4a75,0x5a54,0x6a37,0x7a16,0x0af1,0x1ad0,0x2ab3,0x3a92, 0xfd2e,0xed0f,0xdd6c,0xcd4d,0xbdaa,0xad8b,0x9de8,0x8dc9, 0x7c26,0x6c07,0x5c64,0x4c45,0x3ca2,0x2c83,0x1ce0,0x0cc1, 0xef1f,0xff3e,0xcf5d,0xdf7c,0xaf9b,0xbfba,0x8fd9,0x9ff8, 0x6e17,0x7e36,0x4e55,0x5e74,0x2e93,0x3eb2,0x0ed1,0x1ef0 ); $crc = 0x0000; for ($i = 0; $i < strlen($ptr); $i++) $crc = $crc_table[(($crc>>8) ^ ord($ptr[$i]))] ^ (($crc<<8) & 0x00FFFF); return $crc;}$test = chr(0xC6).chr(0xCE).chr(0xA2).chr(0x03); // CRC16-CCITT = 0xE2B4$key1='key1'; $key2='key2';$key3='key3';echo $key1_db=redisCRC16($key1)%16384; //得出在redis(集群)中键值为'key1'的数据存储插槽为9189echo "<br/>";echo $key2_db=redisCRC16($key2)%16384; //得出在redis(集群)中键值为'key2'的数据存储插槽为4998echo "<br/>";echo $key3_db=redisCRC16($key3)%16384; //得出在redis(集群)中键值为'key3'的数据存储插槽为935根据crc16算法取得的redis数据分区插槽取得 分区所在服务器地址:
首先我们先看一下本地集群环境中的redis分区插槽区间范围(cd 切换到redis安装包目录下的src目录[注意:redis解压包包非安装后的程序所在目录地址]):
[[email protected] src]$ ./redis-trib.rb check 127.0.0.1:6384Connecting to node 127.0.0.1:6384: OKConnecting to node 127.0.0.1:6381: OKConnecting to node 127.0.0.1:6383: OKConnecting to node 127.0.0.1:6379: OKConnecting to node 127.0.0.1:6380: OKConnecting to node 127.0.0.1:6382: OK>>> Performing Cluster Check (using node 127.0.0.1:6384)S: 91329dacb2ac77d9295ed46ecaaec6f2c415f7f6 127.0.0.1:6384 slots: (0 slots) slave replicates 0be8c0b96e23a5843ece9d86cb6d287c92529a8cM: 0be8c0b96e23a5843ece9d86cb6d287c92529a8c 127.0.0.1:6381 slots:10923-16383 (5461 slots) master 1 additional replica(s)S: 53926c1f8b757c6db2d53e12ee94b8c1a761e663 127.0.0.1:6383 slots: (0 slots) slave replicates beff88cb6dcf6897a6c6de36350032984a4bdf33M: 2fb7a8edab2038d7fabe305dd0099de8bdf1f1e6 127.0.0.1:6379 slots:0-5460 (5461 slots) master 1 additional replica(s)M: beff88cb6dcf6897a6c6de36350032984a4bdf33 127.0.0.1:6380 slots:5461-10922 (5462 slots) master 1 additional replica(s)S: 4738074195072ae29c3f3160382e97c3b56a6392 127.0.0.1:6382 slots: (0 slots) slave replicates 2fb7a8edab2038d7fabe305dd0099de8bdf1f1e6[OK] All nodes agree about slots configuration.>>> Check for open slots...>>> Check slots coverage...[OK] All 16384 slots covered.
以上内容可以看到M位主，S为从(官方要求的redis集群环境必须为6台以上偶数形式的服务器数量，否则无法创建集群环境)从上面输出可以看到分区插槽所在服务器位置： slots:10923-16383 (5461 slots) 127.0.0.1:6381 对应从机 地址为127.0.0.1:6384slots:0-5460 (5461 slots) 127.0.0.1:6379 master 对应从机 地址为127.0.0.1:6381slots:5461-10922 (5462 slots) 127.0.0.1:6380 master对应从机 地址为：127.0.0.1:6383
一共三台主分区，可以通过 取得的插槽位置 定位到 不同的redis分区上去取对应的数据，一但主分区数量有变动，就需要根据实际分区数量重新定位插槽分区范围存取redis地址根据ceil(redisCRC16(redis $key值)%16384 / intval(16384/集群master数)) 得到 对应 的 redis集群服务器顺序 地址，并根据对应区间地址指向到对应的 redis区间master地址即可，注意:如果对应的master区间挂掉了，也不必担心，只需要将地址变更为master主区间对应的 slave从区间服务器地址即可一样取得redis数据，而这正是redis3.0后续版本的精妙之处，不需要第三方插件即可在服务端制动实现集群主备模式，不会影响用户正常读取，当然如果是master和对应的slave一起挂掉了那就没办法了，通常这种几率是很小很小的 小到可以忽略即可...关于redis添加删除集群节点的方法请参阅相关资料，摘取天上星 原创，转载请标明作者出处，本文暂不详叙！