Redis (20) a summary of Redis

1 What is Redis

Redis (remote DIctionary server) is an open source memory database that is commonly used as a cache or message queue.

Features of Redis:

• Redis exists in memory, uses hard disk as persistence, and reads and writes 100,000 per second.
• With rich data structure, string, hash table, list, collection, ordered set, provide intersection, set, difference set and other operations.
• Set the TTL to survive and automatically delete the expiration time.
• Redis single-threaded, memcached multithreading, for general application scenarios, single-threaded is enough to use, the advantage is the multi-data type, persistence.
• Data can be copied to any number of slave servers.

2 data structure strings in Redis

• Set KeyName value set key value KeyName
• Keys * Gets all the key values
• Get KeyName Gets the value of KeyName
• EXISTS KeyName Determine if there is a key value KeyName
• DEL keyname Delete key value KeyName
• Type KeyName to determine the types of KeyName (string string, hash hash table, list, set set, Zset ordered set)

Hash table

• Hset KeyName Key value adds a hash attribute and value
• Hget keyname Key Get value
• Hkeys KeyName Get all the key values
• Hvals KeyName Get all the value values
• Hgetall KeyName get all the values

List

Through the two-way linked list implementation, the tail-to Operation is O (1), get the specified element is relatively slow.

• Lpush keyname value1 or Lpush keyname value1 value2 on the left insert
• Rpush keyname value2 Insert on right
• Lrange KeyName 0-1 querying all the elements
• Lpop KeyName
• Rpop KeyName

Collection

Elements are unique, but without order. Using a hash table with an empty value, the operation is O (1).

• Sadd keyname value1 value2 Add value
• Srem KeyName value1 Delete
• Smembers KeyName Querying all elements

Ordered collection

Using hashing and jumping tables, the intermediate speed is also very fast.

• Zadd KeyName Key Value Added
• Zrange keyname start end gets the element of the specified range

Comparison to the list:

Same:

• It's all orderly.
• Can get an element of a range

Different:

• The list is implemented by two-way linked lists, with very fast data access and slow intermediate
• The ordered set is implemented by hashing and jumping tables, and the intermediate speed is also fast, O (log (N))
• A list cannot simply adjust the position of an element, and an ordered collection can
• An ordered set is better than a list to charge memory

3 Redis Persistence

Redis provides two ways to persist: 1 Rdb Snapshot mode 2 aof mode

RDB mode:

When certain conditions are met, a child process is created, the current data is copied, the data is written to a file on the hard disk, and the original storage file is replaced after the write is completed. Data is typically stored in Dump.rdb. On UNIX systems, write-time replication is supported, that is, a persistent write to disk is started, and if other data changes at this time, a copy of the data is executed.

In addition to this automatic snapshot approach, command mode persistence is supported:

• SAVE: Persisted with the parent process in a blocking manner, and no other request could be executed at this time.
• BGSAVE: Persistent through the way of fork subprocess.

AoF Way:

Each operation logs a command, which causes redundancy in some commands, such as adding a property and then deleting it, both of which are redundant. Redis provides some optimizations, so you can avoid these redundant information. Commands are recorded in appendonly.aof.

4 Redis Message Queuing

Redis is used for Message Queuing, and there are usually two ways to use it:

List: All of the consumer data is added to the list, and all of the data in it is listed.

Publish/Subscribe: Each consumer subscribes to a separate channel and each data is independent.

Redis (20) a summary of Redis