Meituan two sides: redis five basic data structures?

Redis 5 Basic data structures （String、List、Hash、Set、Sorted Set） Often asked in an interview , Let's review this article .

Here is the body .

You can Redis Found on the official website Redis The data structure is introduced in great detail ：

• Redis Data Structures[1]

• Redis Data types tutorial[2]

The future as Redis The release of the new version , New data structures may appear , By looking up Redis Introduction to the official website , You can always get the most reliable information .

String（ character string ）

Introduce

String yes Redis The simplest and most commonly used data structure in .

String Is a binary secure data structure , It can be used to store any type of data, such as strings 、 Integers 、 Floating point numbers 、 picture （ The image base64 Encoding or decoding or the path of the picture ）、 Serialized object .

although Redis Yes, it is C Written language , however Redis Not used C String representation of , It's about building a kind of   Simple dynamic string （Simple Dynamic String,SDS）. Compared with C The native string of ,Redis Of SDS You can save not only text data, but also binary data , And the complexity of getting the length of the string is O(1)（C String is O(N)）, besides ,Redis Of SDS API Is safe , No buffer overflow .

Common commands

more Redis String Commands and detailed instructions , Please check out Redis Introduction to the official website ：https://redis.io/commands/?group=string .

Basic operation  ：

> SET key value
OK
> GET key
"value"
> EXISTS key
(integer) 1
> STRLEN key
(integer) 5
> DEL key
(integer) 1
> GET key
(nil)

Batch settings  ：

> MSET key1 value1 key2 value2
OK
> MGET key1 key2 #  Get more than one in batch  key  Corresponding  value
1) "value1"
2) "value2"

Counter （ When the content of the string is an integer, you can use ）：

> SET number 1
OK
> INCR number #  take  key  The value of the number stored in is increased by one 
(integer) 2
> GET number
"2"
> DECR number #  take  key  Subtract one from the number stored in 
(integer) 1
> GET number
"1"

Set expiration time （ The default is never expire ）：

> EXPIRE key 60
(integer) 1
> SETNX key 60 value #  Set the value and set the expiration time 
OK
> TTL key
(integer) 56

Application scenarios

Scenarios that need to store regular data

• give an example ： cache session、token、 Picture address 、 Serialized object ( Compared with Hash Storage saves more memory ).

• Relevant command ：SET、GET.

Scenes that need to be counted

• give an example ： Number of requests per user unit time （ Simple current limiting can be used ）、 Number of page visits per unit time .

• Relevant command ：SET、GET、 INCR、DECR .

Distributed lock

utilize  SETNX key value  Command can realize the simplest distributed lock （ There are some drawbacks , It is generally not recommended to implement distributed locks in this way ）.

List（ list ）

Introduce

Redis Medium List In fact, it is the implementation of linked list data structure . I am here   Linear data structure : Array 、 Linked list 、 Stack 、 queue [3]  This article introduces the data structure of linked list in detail , I won't introduce more here .

Many high-level programming languages have built-in implementation of linked lists, such as Java Medium  LinkedList, however C The language doesn't implement the linked list , therefore Redis It realizes its own linked list data structure .Redis Of List The implementation of is a   Double linked list , That is, it can support reverse lookup and traversal , Easier to operate , But it brings some extra memory overhead .

Common commands

more Redis List Commands and detailed instructions , Please check out Redis Introduction to the official website ：https://redis.io/commands/?group=list .

adopt  RPUSH/LPOP  perhaps  LPUSH/RPOP Implementation queue  ：

> RPUSH myList value1
(integer) 1
> RPUSH myList value2 value3
(integer) 3
> LPOP myList
"value1"
> LRANGE myList 0 1
1) "value2"
2) "value3"
> LRANGE myList 0 -1
1) "value2"
2) "value3"

adopt  RPUSH/RPOP perhaps LPUSH/LPOP  Implementation stack  ：

> RPUSH myList2 value1 value2 value3
(integer) 3
> RPOP myList2 #  take  list The head of ( Far right ) Take out the elements 
"value3"

I drew a special picture for everyone to understand  RPUSH , LPOP , lpush , RPOP  command ：

adopt  LRANGE  View the list elements corresponding to the subscript range  ：

> RPUSH myList value1 value2 value3
(integer) 3
> LRANGE myList 0 1
1) "value1"
2) "value2"
> LRANGE myList 0 -1
1) "value1"
2) "value2"
3) "value3"

adopt  LRANGE  command , You can be based on List Realize paging query , Very high performance ！

adopt  LLEN  Check the list length  ：

> LLEN myList
(integer) 3

Application scenarios

Information flow display

• give an example ： The latest article 、 What's new .

• Relevant command ：LPUSH、LRANGE.

Message queue

Redis List Data structures can be used as message queues , It's just that the function is too simple and has many defects , This is not recommended .

relatively speaking ,Redis 5.0 A new data structure is added  Stream  It is more suitable for message queuing , But the function is still very simple . Compared with professional message queuing , There are still many deficiencies, such as the problem of message loss and accumulation .

Hash（ Hash ）

Introduce

Redis Medium Hash It's a String Type of field-value（ Key value pair ） Mapping table , Ideal for storing objects , When following up , You can directly modify the values of some fields in this object .

Hash Be similar to JDK1.8 Before  HashMap, The internal implementation is similar ( Array + Linked list ). however ,Redis Of Hash More optimizations .

Common commands

more Redis Hash Commands and detailed instructions , Please check out Redis Introduction to the official website ：https://redis.io/commands/?group=hash .

Analog object data storage  ：

> HMSET userInfoKey name "guide" description "dev" age "24"
OK
> HEXISTS userInfoKey name #  see  key  Corresponding  value Whether the field specified in the .
(integer) 1
> HGET userInfoKey name #  Gets the value of the specified field stored in the hash table .
"guide"
> HGET userInfoKey age
"24"
> HGETALL userInfoKey #  Gets the specified in the hash table  key  All fields and values of 
1) "name"
2) "guide"
3) "description"
4) "dev"
5) "age"
6) "24"
> HSET userInfoKey name "GuideGeGe"
> HGET userInfoKey name
"GuideGeGe"

Application scenarios

Object data storage scenario

• give an example ： User information 、 Commodity information 、 The article information 、 Shopping cart information .

• Relevant command ：HSET （ Set the value of a single field ）、HMSET（ Set values for multiple fields ）、HGET（ Get the value of a single field ）、HMGET（ Get the value of multiple fields ）.

Set（ aggregate ）

Introduce

Redis Medium Set A type is an unordered set , The elements in the set have no order, but they are all unique , It's kind of like Java Medium  HashSet . When you need to store a list of data , You don't want duplicate data ,Set Is a good choice , also Set It provides a way to judge whether an element is in a Set Important interfaces within a collection , This is also List What cannot be provided .

You can be based on Set Easy intersection 、 Combine 、 Operation of difference set , For example, you can store all the followers of a user in a collection , Put all its fans in a collection . In this case ,Set Can be very convenient to achieve such as common concern 、 Common fans 、 Common preferences and other functions . This process is also the process of intersection .

Common commands

more Redis Set Commands and detailed instructions , Please check out Redis Introduction to the official website ：https://redis.io/commands/?group=set .

Basic operation  ：

> SADD mySet value1 value2
(integer) 2
> SADD mySet value1 #  Duplicate elements are not allowed , So add failed 
(integer) 0
> SMEMBERS mySet
1) "value1"
2) "value2"
> SCARD mySet
(integer) 2
> SISMEMBER mySet value1
(integer) 1
> SADD mySet2 value2 value3
(integer) 2

• mySet : value1、value2 .

• mySet2 ：value2 、value3 .

Find the intersection  ：

> SINTERSTORE mySet3 mySet mySet2
(integer) 1
> SMEMBERS mySet3
1) "value2"

Union  ：

> SUNION mySet mySet2
1) "value3"
2) "value2"
3) "value1"

Difference set  ：

> SDIFF mySet mySet2 #  The difference set is composed of all that belong to  mySet  But that does not belong to  A  A collection of elements of 
1) "value1"

Application scenarios

Scenarios where the data to be stored cannot be duplicated

• give an example ： Website UV Statistics （ The scenario with a huge amount of data is still  HyperLogLog More suitable for ）、 I like the article 、 Dynamic likes and other scenes .

• Relevant command ：SCARD（ Get the number of sets ） .

You need to get the intersection of multiple data sources 、 Union and difference scenes

• give an example ： Common friends ( intersection )、 Common fans ( intersection )、 Pay close attention to ( intersection )、 Friend recommendation （ Difference set ）、 Music to recommend （ Difference set ） 、 Subscription number recommendation （ Difference set + intersection ） Such as the scene .

• Relevant command ：SINTER（ intersection ）、SINTERSTORE （ intersection ）、SUNION （ Combine ）、SUNIONSTORE（ Combine ）、SDIFF（ intersection ）、SDIFFSTORE （ intersection ）.

Scenarios that require random access to elements in the data source

• give an example ： The lottery system 、 Random .

• Relevant command ：SPOP（ Randomly get the elements in the set and remove , It is suitable for scenes where repeated winning is not allowed ）、SRANDMEMBER（ Get the elements in the collection at random , Suitable for scenarios that allow repeated winning ）.

Sorted Set（ Ordered set ）

Introduce

Sorted Set Be similar to Set, But and Set comparison ,Sorted Set Added a weight parameter  score, To enable elements in a collection to press  score  Arrange in order , You can also use  score  To get a list of elements . It's kind of like Java in  HashMap  and  TreeSet  The combination of .

Common commands

more Redis Sorted Set Commands and detailed instructions , Please check out Redis Introduction to the official website ：https://redis.io/commands/?group=sorted-set .

Basic operation  ：

> ZADD myZset 2.0 value1 1.0 value2
(integer) 2
> ZCARD myZset
2
> ZSCORE myZset value1
2.0
> ZRANGE myZset 0 1
1) "value2"
2) "value1"
> ZREVRANGE myZset 0 1
1) "value1"
2) "value2"
> ZADD myZset2 4.0 value2 3.0 value3
(integer) 2

• myZset : value1(2.0)、value2(1.0) .

• myZset2 ：value2 （4.0）、value3(3.0) .

Get the rank of the specified element  ：

> ZREVRANK myZset value1
0
> ZREVRANK myZset value2
1

Find the intersection  ：

> ZINTERSTORE myZset3 2 myZset myZset2
1
> ZRANGE myZset3 0 1 WITHSCORES
value2
5

Union  ：

> ZUNIONSTORE myZset4 2 myZset myZset2
3
> ZRANGE myZset4 0 2 WITHSCORES
value1
2
value3
3
value2
5

Difference set  ：

> ZDIFF 2 myZset myZset2 WITHSCORES
value1
2

Application scenarios

It is necessary to randomly obtain the elements in the data source and sort them according to a certain weight

• give an example ： Various leaderboards, such as the leaderboard for gifts in the live studio 、 Wechat ranking of friends 、 Ranking list of the ranks in the glory of the king 、 Topic heat ranking list and so on .

• Relevant command ：ZRANGE ( Sort from small to large ) 、 ZREVRANGE （ Sort from large to small ）、ZREVRANK ( Specify the element ranking ).

Scenarios where the data to be stored has priority or importance   For example, priority task queue .

• give an example ： Priority task queue .

• Relevant command ：ZRANGE ( Sort from small to large ) 、 ZREVRANGE （ Sort from large to small ）、ZREVRANK ( Specify the element ranking ).

If this article helps you , Don't forget to give me a 3 even , give the thumbs-up , forward , Comment on ,

Learn more JAVA Knowledge and skills , Pay attention to Blogger learning JAVA Courseware , Source code , Installation package , There are also the latest interview materials of big factories, etc

I'll see you next time .

Collection It's like whoring for nothing , Praise is the truth .