Redis learning - 01 introduction, installation and configuration

Redis brief introduction

Redis The origin of

2008 year , A start-up in Italy Merzia Launched a model based on MySQL Web site real time statistics system based on LLOOGG, But it didn't take long , The founder of the company Salvatore Sanfilippo Then the MySQL Disappointed with the performance of , So he decided to do it himself LLOOGG Customize a database , And in 2009 Development completed in , This database is Redis.

however Salvatore Sanfilippo It's not enough to just Redis be used for LLOOGG This product , But I want more people to use it , So in the same year Salvatore Sanfilippo take Redis Open source release , And start and Redis Another major code contributor to Pieter Noordhuis Go on together Redis Development of , Until today, .

Salvatore Sanfilippo I didn't think of , In just a few years ,Redis We have a huge user base . At home, such as BAT、 Sina weibo 、 Knowledge and so on , Abroad is like GitHub、Stack Overflow、Flickr Are all Redis Users of .

Development timeline ：

• VMware Company from 2010 Started sponsoring Redis Development of ,Salvatore Sanfilippo and Pieter Noordhuis Also in the same year 3 The month and 5 Month join VMware, Full time development Redis

• and 2013 year 5 Month to 2015 year 6 During the month , It was developed by Pivotal sponsorship

• 2015 year 6 month ,Redis Labs Sponsor for further development until now

Redis Code is hosted in GitHub On （https://github.com/redis/redis）, Development is very active .

What is? Redis

The following is the official document explanation .

Redis It's a use ANSI C Write open source 、 Support network 、 Memory based 、 Optional persistent key value pair to store database .

It can meet the storage requirements of different scenarios by providing a variety of key data types , for example ：

• string - character string
• hash - Hash
• list - list
• set - aggregate
• Sort set with range query
• bitmap - Bitmap
• hyperlogolog - Super log
• geospatial index - Geospatial index
• stream - flow

Redis With built-in replication function , Parsing performed Lua Script ,LRU Cache control , Transactions and different levels of disk persistence , And pass Redist Sentinel and Redis Cluster Automatic partitioning provides high availability .

Redis Storage structure of

There is a data structure in most programming languages ： Dictionaries , For example code dict["key"] = "value" in ：

• dict Is a dictionary structure variable
• key Is the key name
• value Is the key value

In the dictionary, we can get or set the key value corresponding to the key name , You can also delete a key .

Redis yes REmote DIctionary Server（ Remote dictionary server ） Abbreviation , It stores data in a dictionary structure , And allow other applications to pass through TCP The protocol reads and writes the contents of the dictionary .

Redis The key value in the dictionary can be a string , It can also be other data types . The more common ones are ：

type	explain
Strings	character string
Hashes	Hash / hash , Is the content that consists of fields associated with values . Fields and values are strings . This is related to Ruby or Python Hashes are very similar . Be similar to JavaScript Object structure in .
Lists	list , A collection of string elements sorted by insertion order . They are basically linked lists .
Sets	A collection of unordered string elements , The data in the set is not duplicated
Sorted sets	And Sets similar , But each string element is associated with a floating-point value called a fraction . Elements are always sorted by their scores , Therefore, Sets Different , You can retrieve a series of elements （ for example , You may ask ： Before you give it to me 10 First or last 10 name ）

Memory storage and persistence

Redis All data in the database is stored in memory . Relative to disk , Memory data read / Write much faster , So we usually use Redis Do cache database , On an ordinary computer ,Redis Can read and write more than... In a second 10 Ten thousand key values .

Redis The performance test on the official website shows that , stay Linux 2.6、Xeon X3320 2.5 GHz Server ,50 In the case of concurrent requests 100000 Time ,SET Operation can reach 110000 Time /s,GET Operation can reach 81000 Time /s

There is also a problem storing data in memory , For example, the data in the memory will be lost after the program exits . however Redis Provides support for persistence , That is, the data in memory can be asynchronously written to the hard disk , At the same time, it does not affect the continued service .

Rich in functions

Redis Although it was developed as a database , But because it provides a wealth of functions , More and more people use it as a cache 、 Queue system, etc .

1、 As a caching system

Redis You can set the lifetime for each key , When the lifetime expires, it will be deleted automatically . This feature, combined with excellent performance, makes Redis Can be used as a cache . As a caching system ,Redis It can also limit the maximum space occupied by data , After the data reaches the space limit, the unnecessary keys can be eliminated automatically according to certain rules .

2、 As a queue system

besides ,Redis The list type key of can be used to implement the queue , It also supports blocking reading , It is easy to implement a high-performance priority queue .

3、“ Release / subscribe ” function

At the same time on a higher level ,Redis And support “ Release / subscribe ” Message mode for , You can build chat rooms and other systems based on this .

Simple and stable

Even if the function is rich , It is difficult to attract people if it is too complex to use .Redis The intuitive storage structure makes it possible to Redis The interaction is very simple .

stay Redis Use command to read and write data , A command statement is to Redis Equivalent to SQL Language for relational databases . For example, in a relational database, you want to get posts In table id by 1 The record of title Fields can be used as follows SQL Statements for ：

SELECT title FROM posts WHERE id=1 LIMIT 1

Corresponding , stay Redis The name of the key to be read in is post: 1 Hash type of key title Value of field , You can use the following statement to implement ：

HGET post:1 title

among ,HGET It's an order ,post:1 Is the key name ,title Is the data field to be read .

Redis Provides 250 Multiple orders , It sounds a lot , But there are only a few dozen commonly used , And every command is easy to remember .Redis The list of commands ：Command reference – Redis

Redis Provides clients in dozens of different programming languages （https://redis.io/clients）, These libraries are well encapsulated Redis The order of , Make in a program with Redis It's easier to interact .

Redis Application scenarios

Redis It's a Key-Value The storage system , In most cases, it is because of its high-performance characteristics , Used as a cache , Here's an introduction Redis Frequently encountered usage scenarios .

The use scenario of a product must be based on the characteristics of the product , Let's start with a list of Redis Characteristics ：

• Excellent reading and writing performance
• Persistence
• Rich data types
• Single thread
• Data automatically expires
• Publish subscribe
• Distributed

Here we go through several scenarios , Different dimensions Redis Application .

1、 Cache system

Caching is now used by almost all large and medium-sized websites , The proper use of caching can not only improve the speed of website access , It also greatly reduces database stress .Redis Key expiration is provided , It also provides a flexible key elimination strategy , therefore , Now? Redis It's used a lot for caching .

2、 Ranking List

Many websites have leaderboard applications , Such as Jingdong's monthly sales list 、 The new ranking of goods by time, etc .Redis The provided ordered collection data class can realize all kinds of complex leaderboard applications .

3、 Counter

What is a counter , For example, the number of visits to products on e-commerce websites 、 The number of videos played on video websites, etc . In order to ensure that the data is effective , Every time you browse, you have to give +1, When the amount of concurrency is high, it is undoubtedly a challenge and pressure to request database operations every time .Redis Provided incr To implement the counter function , Memory operations , Very good performance , Great for these counting scenarios .

4、 Distributed session

Cluster mode , In the case of few applications, it is generally used that the container comes with session Copy function can satisfy , As applications increase, in relatively complex systems , It's usually built to Redis And so on session service ,session No longer managed by containers , But by the session Service and memory database management .

5、 Distributed lock

Distributed technology is used in many Internet companies , The technical challenge of distributed technology is concurrent access to the same resource , Such as the overall situation ID、 Reduce inventory 、 Second kill and other scenes , The pessimistic lock of the database can be used in the scenario with small amount of concurrency 、 Optimistic lock to achieve , But in the case of high concurrency , It is not ideal to use database lock to control concurrent access of resources , Greatly affected the performance of the database . You can use Redis Of setnx Function to write distributed locks , If set, return to 1 Indicates that the lock is obtained successfully , Otherwise, the lock acquisition fails , There are more details to consider in practical application .

6、 Social networks

give the thumbs-up 、 Step on 、 Focus on / Be focus on 、 Mutual friends and so on are the basic functions of social networking sites , Social networking sites tend to get a lot of traffic , And traditional relational database types are not suitable for storing this type of data ,Redis Provided hash 、 Data structures such as collections can easily implement these functions .

7、 Latest list

Redis List structure ,LPUSH You can insert a content in the head of the list ID As a keyword ,LTRIM Can be used to limit the number of lists , So the list is always N individual ID, No need to query the latest list , Direct basis ID Go to the corresponding content page .

8、 The messaging system

Message queue is a necessary middleware for large websites , Such as ActiveMQ、RabbitMQ、Kafka And other popular Message Queuing Middleware , It is mainly used for business decoupling 、 Peak shaving and asynchronous processing of low real-time services .Redis Provides a release / Subscription and blocking queue function , Can realize a simple message queue system . in addition , This can't be compared with professional message middleware .

Example ： Secshahe Redis The combination of

Seckill is a common marketing mode in the Internet system , As a developer , In fact, I don't like such activities , Because non-technical personnel can not understand the technical difficulty , As a result, there are always some deviations in resource coordination . In fact, the common problems of secsha include ：

• Too high concurrency leads to program blocking .
• Inventory cannot be effectively controlled , Oversold occurs .

In fact, there are basically two solutions to these problems ：

• Try to cache the data , Block the direct interaction between users and the database .
• Lock to avoid oversold .

Now let's explain , If we do a second kill now , that ,Redis How they should be used together ?

• Preheat data in advance , Put in Redis
• Product list put Redis List
• Detailed data of goods Redis Hash preservation , Set expiration time
• Inventory data of goods Redis Sorted Set preservation
• User's address information Redis Set preservation
• The order is generated by deducting inventory Redis Manufacturing distributed locks , Inventory synchronous deduction
• Shipment data after order generation , produce Redis List, Processing through message queuing
• After the second kill , And then Redis Data and database synchronization

The above is a brief second kill system and Redis A combined solution , Of course, it may be introduced in practice HTTP cache , Or use message docking MQ Alternatives, etc , There will also be business omissions , This is just a hope that it can attract jade .

Related resources

• Official website ：https://redis.io/
• GitHub Warehouse ：https://github.com/redis/redis
• Interactive learning Redis（ Set up Redis learning environment ）：https://try.redis.io/
• Redis Chinese net （ unofficial ）：http://www.redis.cn/
• Redis Command Reference ：http://doc.redisfans.com/

Redis install

About Redis Version of

Redis Learn from it Linux The naming rules of operating system for version number ：

• If the second digit of version number is an odd number , It's an unstable version （Unstable）, for example 2.7、2.9、3.1
• If the second digit of the version number is even , The stable version （Stable）, for example 2.6、3.0、3.2

The current odd version is the next stable development version , for example 2.9 The version is 3.0 Development version of version .

Even versions of... Are usually selected in the production environment Redis, The odd version is used to learn and experience some new features in advance .

obtain Redis The way

obtain Redis There are many ways ：

• Install on your own computer
• Install on virtual machine
• Install to the remote server
• from Docker Hub obtain Redis Of Docker Mirror image
• …

stay macOS Install in Redis

stay macOS There are two ways ：

• Mode one ： Compilation and installation , Same as Linux
• Mode two （ recommend ）： Use Homebrew install

macOS Software package management tools under the system Homebrew A newer version of Redis package , You can use it directly to install Redis, There is no need for Linux The trouble of manual compilation on .

1、 install Homebrew

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

2、 adopt Homebrew install Redis

brew install redis

stay Windows Install in Redis

Redis The government does not support Windows.

2011 Microsoft direction Redis Submitted a patch , In order to make Redis Can be in Windows Under the compiler run . But it was rejected by the author , The reason is in the server field Linux It has been widely used , Give Way Redis Can be in Windows It is not so important to run under . also Redis There are many operating system related features , compatible Windows It will consume too much energy and affect Redis Development of .

For all that , Microsoft has released a software that can be used in Windows Under the Redis edition , But this project is no longer maintained .

If you really want to Windows Learn to use Redis, You can try Memurai, It's a Redis for Windows substitute , Its core is based on Redis Source code and fully compatible Redis, But the project has not been officially recognized by Microsoft .

stay Linux Install in Redis

• Redis Official installation documentation
• Redis Quick Start

Use the following command to download 、 Extract and compile Redis：

#  download  Redis  Source compression package 
wget https://download.redis.io/releases/redis-6.2.6.tar.gz
#  Or use this address to download , This address always points to the latest stable version 
# wget http://download.redis.io/redis-stable.tar.gz

#  Decompress the package 
tar xzf redis-6.2.6.tar.gz

#  Get into  Redis  Source directory 
cd redis-6.2.6

#  Compilation and installation 
make

Now the compiled binary executable can be src Found in the directory . Execute the following file to run Redis：

#  Run with full path  Redis
src/redis-server
# Ctrl+C  stop it  Redis, The following describes the stop command 

If you want to use it directly Redis Command to run Redis The server , You need to copy the executable to a directory in the environment variable , You do not need to add its full path to execute the executable file located in this directory .

This command can be copied manually or executed directly in the source directory ：

make install

This order will Redis Binary executable copied to /usr/local/bin Directory , PATH Environment variables are configured by default /usr/local/bin Catalog .

#  Direct operation  Redis
redis-server

Redis function 、 Stop and connect

function Redis

The compiled Redis Source directory src There will be the following executable files in the folder ：

Executable file	explain
redis-server	Redis The server
redis-cli	And Redis Interactive command line interface client tools
redis-benchmark	Redis Performance testing tools
redis-check-aof	AOF File repair tool
redis-check-rdb	RDB File checking tool （3.0 And earlier redis-check-dump）
redis-sentinel	Sentry mode tool （ Monitoring and failover ）

in addition , Install by compiling the source code , It will also generate a in the root directory of the source code redis.conf Configuration file for .

The most common of these executables is redis-server and redis-cli.

Direct operation redis-server You can start Redis The server ：

redis-server

Redis By default 6379 port , Can pass --port Parameter specifies the boot port ：

redis-server --port 1234

Direct operation redis-server It will run at the front desk Redis service , Occupy the command line window .

have access to --daemonize Parameter and set to yes, To run on the back end Redis：

redis-server --daemonize yes

see Redis Running state ：

#  View the current contents  redis  The process of 
#  If you can see it  redis-server  The process of , Express  Redis  The server is running 
ps -ef | grep -i redis

stop it Redis

in consideration of Redis It is possible that data in memory is being synchronized to the hard disk , force finish Redis process （ for example Ctrl + C） May cause data loss .

So stop correctly Redis The way is to Redis send out SHUTDOWN command ：

#  Be careful ： It's used here  redis-cli  instead of  redis-server
redis-cli shutdown

When Redis perform SHUTDOWN After the command , Disconnect all clients first , Then perform persistence according to the configuration , Finally, complete the launch .

Redis Can be handled properly SIGTERM The signal , So use kill Redis Process PID It can also end normally Redis, Effect and delivery SHUTDOWN command ：

#  Stop by process number  Redis
kill -9 280697

Can pass ps -ef | grep -i redis Check to see if there are any after the stop redis-server process .

Connect Redis

redis-cli yes Redis Built in command line based Redis Client tools , It is also learning and testing Redis An important tool for .

function redis-cli You can connect to the database ：

redis-cli

Connect by default 127.0.0.1:6379 The database of , You can also specify the server address and port ：

redis-cli -h 127.0.0.1 -p 1234

Under normal circumstances , At this time, it is connected Redis database , You can execute the following test command ：

# PING  If the command outputs  PONG  Indicates that the connection is normal 
127.0.0.1:6379> PING
PONG
#  towards  Redis  Store a  key  by  foo, The value is  bar  The data of 
127.0.0.1:6379> SET foo bar
OK
#  Reading data 
127.0.0.1:6379> GET foo
"bar"
#  Query all data  key
127.0.0.1:6379> KEYS *
1) "foo"

disconnect ：

• command ：quit
• Shortcut key ：Ctrl + C

Redis To configure

As described above, you can use redis-server Start parameter of --port Set up Redis The port number of the service , besides Redis Other configuration options are also supported , For example, whether to enable persistence 、 Log level, etc .

There are several ways to Redis To configure ：

1. Command line arguments
2. The configuration file
3. Change while the server is running Redis To configure

Passing parameters through the command line

The simplest way is to start redis-server Directly pass the command parameters ：

redis-server --port 6379 --host 127.0.0.1

The configuration file

There are many options that can be configured , It is not convenient to set these options through startup parameters , therefore Redis It supports setting these options through the configuration file .

Redis Provides a configuration file modular redis.conf, Located in the root path of the source directory .

This file is just a template for a configuration file , The way to make the configuration effective is to pass the path of the configuration file as a startup parameter to redis-server：

redis-server < Profile path >

Passing a configuration option with the same name through the startup parameter will overwrite the corresponding parameter in the configuration file , for example ：

redis-server < Configure as read path > --port 3000

It is recommended to put the configuration file in a unified directory （ for example /etc/redis）, Named after port number （ for example 6379.conf）.

Copy the configuration file module to the target file ：

cp <redis  Source code root >/redis.conf <redis  Profile directory >/6379.conf

Change while the server is running Redis To configure

You can also do it in Redis Run through redis-cli Of CONFIG SET The command does not restart Redis Dynamic modification in the case of part Redis To configure ：

CONFIG SET logLevel warning

Be careful ： During operation, only part Modify the configuration item .

Again , It can also be used at runtime CONFIG GET Command to get Redis Current configuration ：

CONFIG GET logLevel

Redis Multiple databases in

adopt redis-server Start a Redis Service is a Redis example .

One Redis The example provides several dictionaries for storing data , The client can specify which dictionary to store the data in . This is similar to the well-known idea that multiple databases can be created in a relational database , So each dictionary can be understood as a separate database .

Redis The default support 16 A database , They are numbered ：0、1、2、…、14、15

• Reids Custom database names are not supported
• Because each database is named after a number , So developers must be clear about which database stores which data
• Parameters can be configured by databases Modify the number of supported databases

Each database is independent , That is to say 0 Data inserted into database No , stay 1 No. database is inaccessible .

The client and Redis Automatically select after the connection is established 0 The database , have access to SELECT Command to switch the database ：

#  Connect  Redis  Default connected  0  The database 
redis-cli
127.0.0.1:6379> SET foo bar
OK
127.0.0.1:6379> GET foo
"bar"
127.0.0.1:6379> SELECT 1
OK
# 1  No. database does not store  foo
127.0.0.1:6379> GET foo
(nil)
127.0.0.1:6379> KEYS *
(empty array)
#  The switched database number exceeds the set value , You're going to report a mistake 
127.0.0.1:6379> SELECT 16
(error) ERR DB index is out of range

Redis Setting different access passwords for each database is not supported , So a client either has access to all databases , Or none of them .

Multiple databases are not completely isolated , for example FLUSHALL The command can empty a Redis Data in all databases in the instance .

in summary , These databases are more like a namespace , and Not suitable for storing data from different applications , For example, it is not suitable to use 0 No. database storage A Application data , While using 1 No. database storage B Application data , This is not recommended ！！

Different applications should use different Redis Instance storage data .

because Redis Very light weight , An empty one Redis The memory occupied is only 1 MB about , So don't worry about more than one Redis Instances will take up a lot of extra memory .