Ten common application scenarios of redis

One 、 cache
As Key-Value Form of memory database ,Redis The first application scenario that will be thought of is as data cache . At present, this is a must kill technique 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 , While using Redis Caching data is also very simple , Just go through string type Save the serialized object , But there are some things to pay attention to ：
（ One ） Different objects must be guaranteed key No repetition , And make key Try to be as short as possible , Generally use the class name （ Table name ） It is spliced with a primary key .
（ Two ） Choosing a good serialization method is also important , The purpose is to improve the efficiency of serialization and reduce memory consumption .
（ 3、 ... and ） Consistency between cache content and database , There are generally two approaches here ：
① Only put the object into the cache after the database query , If the object is modified or deleted , Directly clear the corresponding cache （ Or set to expire ）.
② After adding and querying the database, put the object into the cache , Update cache after modification , Clear the corresponding cache after deletion （ Or set to expire ）.

Two 、 Ranking List
Many websites currently have leaderboard applications , Such as the year of Taobao / Daily sales list 、 The new ranking of goods by time, etc . utilize Redis Of zset The structure can realize various complex leaderboard applications . For example, use zset And an algorithm to calculate the heat can easily create a heat ranking list ,zrevrangebyscore You can get the sequence in reverse order of fractions ,zrank You can get a member's position in the leaderboard （ Is the position when the fractions are arranged in positive order , If you want to get the position in reverse order, you need to use zcard-zrank）.

3、 ... and 、 Counter
What is a counter , Such as the reading volume of the article 、 Microblog likes 、 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 . We can write Redis Then synchronize to the database regularly ,Redis Provided incr To implement the counter function , Memory operations , Very good performance , Great for these counting scenarios .
The counting function should be most suitable for Redis One of the usage scenarios of , Because its high-frequency reading and writing characteristics can be brought into full play Redis As an efficient memory database . stay Redis In the data structure of ,string、hash and sorted set provide incr Method is used for atomic self increasing operation , Here are some examples of their respective usage scenarios ：
（1） If the application needs to display the number of registered users per day , You can use string As a counter , Set a name REGISTERED_COUNT_TODAY Of key, And set it to the early morning during initialization 0 The expiration time of the point , Every time the user registers successfully, it will use incr The command makes the key growth 1, At the same time every morning 0 After the point , Because of this key After expiration, the value is cleared .
（2） Every microblog has some likes 、 comments 、 Forwarding number and browsing number are four attributes , Use hash It would be better to count , Set the... Of this counter key Set to weibo:weibo_id,hash Of field by like_number、comment_number、forward_number and view_number, After the corresponding operation, pass hincrby send hash Medium field Self increasing .
（3） If the app has a function of posting leaderboards , choice sorted set More appropriate , Set the key Set to POST_RANK. When users post , Use zincrby Put the user id Of score growth 1.sorted set Will be reordered , The position of users in the leaderboard will be updated in real time .

Four 、 Current limiting
In some seckill activities , In an instant, there will be a large number of user requests pouring in , And probably the same user will constantly refresh the page or other operations to send a large number of requests , At this time, we can take current limiting measures . How to realize current limitation ？ We can use Redis Of incr Method , With the visitor's ip And other information as key, Add one count at a time , When the number of visits of the same user exceeds our preset number, a prompt message will be returned （ For example, prompt the user to operate too frequently , Operate again after a certain time ）.

5、 ... and 、 Distributed session
Cluster mode , In the case of few applications, it is generally used that the container comes with session Copy function can satisfy , When used in more and more 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 .

6、 ... and 、 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 will be more details to consider in practical application .
stay Redis 2.6.12 Version start ,string Of set The command adds some parameters ：
（ One ）EX： Set the expiration time of the key （ The unit is in seconds ）
（ Two ）PX： Set the expiration time of the key （ The unit is millisecond ）
（ 3、 ... and ）NX ： Only if the bond doesn't exist , To set the key . SET key value NX The effect is equivalent to SETNX key value .
（ Four ）XX ： Only if the bond already exists , To set the key .

Because this operation is atomic , You can simply implement a distributed lock , for example ：

set lock_key locked NX EX 1 

If this operation returns false, explain key Failed to add , That is, someone is currently occupying this lock . And if you go back true, It means you have the lock , Then you can continue the operation , And pass... After operation del Command to release the lock . And even if the program does not release the lock for some reason , Due to the expiration time , The lock will also be in 1 Automatically release in seconds , Will not affect the operation of other programs .

Here we recommend you to use redisson Third party libraries to implement distributed locks .

7、 ... and 、 Message queue
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 Redis in list The data structure implementation of is a two-way linked list , So it can be easily applied to message queue （ producer / Consumer model ）. The producer of the message only needs to pass lpush Put the message into list, Consumers can use rpop Take out the message , And it can ensure the order of messages . If you need to implement message queuing with priority, you can also choose sorted set. and pub/sub Features can also be used as publishers / Subscriber model messages . No matter what method is used , because Redis With persistence , There is no need to worry about message loss due to server failure .
List Two blocked pop-up operations are provided ：blpop/brpop, Timeout can be set :
blpop：blpop key1 timeout Remove and get the first element of the list , If there are no elements in the list, it will block the list until the wait timeout or pop-up elements are found .
brpop：brpop key1 timeout Remove and get the last element of the list , If there are no elements in the list, it will block the list until the wait timeout or pop-up elements are found .
The above operation is actually java Blocking queue
queue ： First in, first out ：rpush blpop, Left head right tail , Right into the queue , Queue left
Stack ： First in, then out ：rpush brpop

8、 ... and 、 Luck draw
utilize set Disorder of structure , adopt Spop（ Redis Spop The command is used to remove the specified key One or more random elements of , After removal, the removed element will be returned . ） Random values .

redis> SADD myset "one"
(integer) 1
redis> SADD myset "two"
(integer) 1
redis> SADD myset "three"
(integer) 1
redis> SPOP myset
"one"
redis> SMEMBERS myset
1) "three"
2) "two"
redis> SADD myset "four"
(integer) 1
redis> SADD myset "five"
(integer) 1
redis> SPOP myset 3
1) "five"
2) "four"
3) "two"
redis> SMEMBERS myset
1) "three"
redis> 

Nine 、 give the thumbs-up 、 Sign in
Suppose a user comments , We want to give him some praise , His ID yes t1001, user ID yes u3001.
Then we can use like:t1001 To maintain all like users of this comment .

// Like this Weibo 
sadd like:t1001 u3001
// Cancel likes 
srem like:t1001 u3001
// Do you like it 
sismember like:t1001 u3001
// All users who like 
smembers like:t1001
// Number of likes 
scard like:t1001

This is not simpler than querying the database directly 、 More efficient ？

Ten 、 Show the latest list of projects
for instance , When we want to list the latest users on the web 20 comments , We have a... Next to the latest comment “ Show all ” Link to , Click to get more comments .
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 .
That is, every time a new comment is published , We'll put it ID Add to a Redis list . You can limit the length of the list to 1000
LPUSH latest.comments
stay Redis Our latest ID Using resident cache , It's always updated . But we have made a limit not to exceed 1000 individual ID, So our acquisition ID The function will always ask Redis. Only when it goes beyond this range , To access the database .