What are the principles and usage of Redis type and encoding encoding?

摘要

Redis是一款开源的高性能key-value数据库，广泛应用于各种场景。在Redis中，数据类型（type）和编码（encoding） 是非常重要的概念。本篇博客将详细介绍Redis支持的数据类型以及相应的编码方式和底层实现原理。

要查看Redis某个key的内部编码，可以使用Redis命令OBJECT ENCODING key。其中，key是你想要查询的键名。例如，如果你想要查询名为mykey的键的内部编码，可以执行以下命令：

127.0.0.1:6379> object encoding mykey  // 查看某个Redis键值的编码

redisObject

在 Redis 中，redisObject 是 Redis 中最基本的数据结构之一。redisObject 用于表示 Redis 中的键值对中的值，它可以是字符串、整数、列表、哈希表等任意一种 Redis 数据类型。

redisObject 的定义如下：

typedef struct redisObject {

    // 类型
    unsigned type:4;

    // 编码方式
    unsigned encoding:4;

    // 引用计数
    int refcount;

    // 指向实际值的指针
    void *ptr;

} robj;

• type：表示 redisObject 的类型。

• encoding：表示 redisObject 的编码方式。

• refcount：表示当前 redisObject 被引用的次数。

• ptr： ptr字段则是一个指针，指向实际的 Redis 对象。

Redis源码encoding取值有如下几种：

#define OBJ_ENCODING_RAW 0        /* Raw representation */
#define OBJ_ENCODING_INT 1        /* Encoded as integer */
#define OBJ_ENCODING_HT 2         /* Encoded as hash table */
#define OBJ_ENCODING_ZIPMAP 3     /* Encoded as zipmap */
#define OBJ_ENCODING_LINKEDLIST 4 /* No longer used: old list encoding. */
#define OBJ_ENCODING_ZIPLIST 5    /* Encoded as ziplist */
#define OBJ_ENCODING_INTSET 6     /* Encoded as intset */
#define OBJ_ENCODING_SKIPLIST 7   /* Encoded as skiplist */
#define OBJ_ENCODING_EMBSTR 8     /* Embedded sds string encoding */
#define OBJ_ENCODING_QUICKLIST 9  /* Encoded as linked list of ziplists */

类型与编码介绍

Redis支持五种主要的数据类型：字符串(string)、列表(list)、集合(set)、有序集合(sorted set)和哈希(hash)。每种数据类型都有对应的编码方式。

数据类型与编码方式总览如下：

数据类型	编码方式
字符串	int、embstr、raw
哈希表	ziplist、hashtable
列表	ziplist、linkedlist、quicklist
集合	intset、hashtable
有序集合	ziplist、skiplist

String

String is the most basic data type in Redis, usually used to store text or binary data. Redis supports two encoding methods:

• In Redis, when a string can be represented as an integer, it will be converted to an integer and stored using int encoding. The advantages of int encoding are small storage space and high operating efficiency. The disadvantage is that it can only store integers and does not support string operations.

• embstr(embstr-encoded string): Save a string with a length less than 44 bytes. When a string is relatively short, use this encoding method to store it. Reduce memory usage.

• raw(raw-encoded string): Save strings longer than 44 bytes. When a string is relatively long, this encoding method is used to store it.

List

A list is an ordered collection of strings in which elements can be added, modified, and deleted. Redis supports three encoding methods:

• ziplist: Before Redis3.2 version, When the length of each string in the List is less than 64 bytes and the length of the string in the List is When the number of elements is less than 512, the List object uses ziplist encoding, and in other cases, linkedlist encoding is used. ziplist is a compact, compressed list structure that saves memory. Good for small lists.

• Linkedlist is a type of data structure that supports lists of any size through a linked list structure.. But its memory footprint will increase as the length of the list increases.

• quicklist: Introduced in Redis 3.2 version, quicklist is a list structure composed of multiple ziplists, which can not only ensure performance, but also save memory. Suitable for large lists.

Collection

A collection is a series of unordered string collections that support adding, deleting and querying elements. Redis supports two encoding methods:

• intset: When the elements in the collection are all integers, Redis will use intset encoding to store . The advantages of intset encoding are small storage space and high operating efficiency.

• hashtable: When the elements in the collection contain strings, Redis will use hashtable encoding to store . The advantage of hashtable encoding is that it can store elements of any type and supports string operations. The disadvantage is that the storage space is relatively large and the operating efficiency is relatively low.

Ordered Set

Unordered collection of strings will be assigned a score, so that they can be sorted according to the score and form an ordered set. Redis supports two encoding methods:

• ziplist: Save less than 128 elements and the size of all elements is less than 64 bytes. Use ziplist encoding , ziplist is a A compact, compressed list structure suitable for small ordered collections.

• skiplist: skiplist is a skip list structure that supports fast query and sorting. Suitable for large ordered collections.

Hash table

A hash table is a collection of key-value pairs, with each key associated with a value. Redis supports two encoding methods:

• ziplist: The string length of all key values ​​stored in the hash object is less than 64 bytes and the number of key-value pairs is less than 512, Redis will use ziplist encoding to store. The advantages of the ziplist encoding method are small storage space and high operating efficiency. The disadvantage is that fast key lookup operations are not supported.

• In addition to the above conditions, Redis will use hash table encoding to store. The advantage of hashtable encoding is that it supports fast key lookup operations. The disadvantage is that the storage space is relatively large and the operating efficiency is relatively low.

The underlying principles of types and encoding

After understanding the data types and encoding methods supported by Redis, let's take a look at their underlying implementation principles.

Encoding conversion

In Redis, each key-value pair has a type identifier, which is used to indicate the data type stored in the key-value pair. When we operate on a key, Redis will encode the key-value pair based on the current encoding of the key and the encoding required for the operation.

For example, when we append content to a string, if the current encoding method of the string is raw, but the new content can be stored using embstr encoding, then Redis will encode the string The method is converted from raw to embstr.

Data structure

Redis uses a variety of classic data structures to implement various data types, not just limited to encoding methods. For example, Redis lists and hash tables are implemented using linked list structures. Skip List is an efficient data structure used in ordered collections.

These data structures have been carefully designed and optimized to meet application needs in various scenarios. For example, the linked list structure is suitable for situations where elements need to be frequently added and deleted, while the skip list structure is more suitable for sorting and search operations.

The above is the detailed content of What are the principles and usage of Redis type and encoding encoding?. For more information, please follow other related articles on the PHP Chinese website!