[redis] ② redis general command; Why is redis so fast?; Redis data type

One 、Redis General Command

Connect Redis client

./redis-cli
./redis-cli -a zgqyyes

• keys *： View all key（ Will block other operations , Not recommended ）

keys *

• set key value： Set up the data

set name qingyi

• dbsize： see key The number of 【 Very fast performance , Time complexity O(1)】

dbsize

• exists key： Determine if there is a counterpart key（ There is returned 1, Otherwise return to 0）

exists name

• expire key seconds： To assign to key Set expiration time （ Company ： second ）

#  Set successfully returned  1
# 5s  after  name  This  key  Will be deleted 
expire name 5

• ttl key： see key How long do you have to live （-1： Permanent validity ;-2： Invalid key ）

• persist key： Put a with an expiration date key Change to a permanent key
  
• del key： Delete specified key, You can delete more than one at a time

#  Delete two  key: name  and  sex（ The return value is deleted  key  The number of ） 
del name sex

• type key： see key Data type of

type name

• flushdb and flushall： Clear database records
• info: View server information

Two 、 Why? Redis It's going to be so fast ？

① Redis Of data yes Stored in memory Of
② Data processing model
③ Avoid the race consumption of context switching （ Single process ）
④ Use epoll Model , Non blocking IO
⑤ Redis It's not a single thread （ The execution directory is single threaded ）

3、 ... and 、Redis Data type of （*）

Data type refers to Value The type of data stored
The value of the string is the largest 512M
Deposited value Can be a string 、 The number 、 Binary system 、JSON data

(1) String

a. command

value The value of is a string type , It can also be binary data and numbers
command ①：set、get、del

set name pzj
get name
del name

command ②：incr、incrby、decr、decrby（ Atomic operation ）
The value increases or decreases 1
The value increases or decreases the specified value

# age  Of  value  Reduce 2
decrby age 2

command ③：set、setnx、setxx、setex
set： Add or modify （ If it doesn't exist, add , Modification of existence ）
setnx： New operation （ If it exists, it will not be processed ）
setxx： Modify the operating （ Modification of existence , If it does not exist, it will not be processed ）
setex： Set up key The expiration time of

command ④：mget、mset（ Get or set multiple at a time key-value）

command ⑤：getset、append、strlen
getset： Set the value and return the old value
append： String concatenation
strlen： String length

getset name pzj
append name lovezgq
strlen name

command ⑥：incrbyfloat、getrange、setrange
incrbyfloat： Decimal increments , The negative number passed in decreases
getrange： Get some data （ Intercepting string ）
settrange： Replace some data

b. Application scenarios

• Caching function （ The most classic use scenario of string ）：Redis As a cache layer ,MySQL As storage layer . Most of the requested data is in Redis In order to get , because Redis It can support high concurrency , Therefore, caching can usually speed up the reading and writing speed and reduce the back-end pressure

• Counter ： Many applications use Redis As a counting tool . It enables fast counting 、 The function of querying cached data , At the same time, data can be landed to other data sources step by step . （ Such as ： Video playback volume ）

• share Session： For load balancing , Distributed services will balance the access of user information to different servers , Users may need to log in again for a refresh . To avoid this problem, you can use Redis Will the user Session centralized management . In this mode , Just promise Redis High availability and scalability , Every time you get user updates or query login information, you can directly from Redis To obtain

• The speed limit ： On safety grounds , Ask the user to enter the mobile phone verification code every time they log in , For short message interface is not frequently accessed , It will limit the frequency of obtaining the verification code per minute

(2) Hash

Hash The type is String Type of field and value Mapping table , Or a String aggregate . It's especially suitable for Store the object , By comparison , Store an object in Hash Type is better than stored in String Type takes up less memory space .

a. command

command ①：hset、hget、hdel

{
	handsome:boy1: {
		id: 1,
		name: pzj,
		age: 18
	}
}

command ②：hexists（ Determine whether the field exists ）;hlen（ Judge how many field Field ）

command ③：hmget（ Get the value of multiple fields ）;hmset（ Set values for multiple fields ）

command ④：hgetall、hvals、hkeys
hgetall： according to key return hash All fields and values of
hvals： according to key Return all values
hkeys： according to key Return all fields

command ⑤：hsetnx、hincrby、hincrbyfloat
hsetnx： modify field Value of field . If the value exists , No modification ; If it doesn't exist , Modify
hincrby： For fields field Increase or decrease
hincrbyfloat： For fields field Increase or decrease

b. Application scenarios

Hash structure is more intuitive than string serialization cache information , And it's more convenient to update . Often used in User information Management of , but Hash type Different from relational databases , The hash type is sparse Of . And relational databases are completely structured , Relational databases can do complex relational queries , and Redis To simulate complex queries of relational databases , open
It's difficult to send , Maintenance costs are high .

Go to Redis Store user information in ：

① Use string type ： Modify and add Field trouble
② Use Hash type ： You can directly modify a Field Or add a Field

(3) List

It is equivalent to storing array types

• In order
• Can be repeated

Redis in Of List similar Java Medium Queue（ queue ）, It can also be used as List To use
List Type is a set of linked list structures , Its main functions are push、pop 、 Get elements, etc .Redis Of List The type is Double ended linked list Structure , It can be done through correlation api Add or delete elements at the head or tail of the collection .

lpush： Add elements from the head
lrange fruit 0 -1（ From beginning to end ）
rpush： Add elements from the tail
lpop： Remove element from header , And return the deleted element
rpop： Remove element from tail , And return the deleted element

linsert： Go to the front of the specified element （Before） Or behind （After） Insert elements

lset： Replace the element of the specified index

lrem： Remove elements , Returns the number of deleted elements
lindex： The return name is key Of list in index The element of location
llen： Returns the number of elements

ltrim： Keep assignments key Elements within the specified range of are not deleted

(4) Set

command ①：sadd、srem、smembers

command ②：spop
Randomly delete a specified number of elements , And return the value of the deleted element

sdiff： Return different elements in two sets （ That set is the standard in front of that set ）
sdiffstore： Store the different elements returned in another collection
sinter： Returns the intersection of sets
sinterstore： Return the intersection result and store it in another set
sunion： Union and collection
sunionstore： Get Union , Into another set
smove： From a set Move the collection to another set In the assembly （ It's equivalent to cutting and copying ）
scard： View the number of elements in the set
sismember： Determine whether an element is an element in the set
srandmember： Returns a random element

(5) ZSet

zadd： Add an element to an ordered collection （ If the element exists, update the order ）
zrange： Sort according to the score from small to large
zrevrange： Sort according to the score from big to small
zrem： Delete the name as key Of zset The elements in
zincrby： Automatically increase or decrease with the specified value （ Usage and previous incrby type ）
zrangebyscore： Return the data within the specified range
zremrangebyrank： Deletes the element at the specified location
zremrangebysocre： Delete the data of the specified score range
zcard： Returns the number of all elements in the collection
zcount： Go back to the collection score The quantity in a given interval

Ranking List ： Orderly collection of classic use scenarios .
For example, video websites need to rank videos uploaded by users . The maintenance of the list may be multifaceted （ According to the time 、 According to the amount of play 、 According to the number of likes you get ）