Shulou(Shulou.com)06/03 Report--

This article mainly introduces the reasons for the high efficiency of Redis and the example analysis of the data structure, the article introduces in great detail, has a certain reference value, interested friends must read!

1. What is redis? What is it mainly used for?

Redis, the English full name is Remote Dictionary Server (remote Dictionary Service), is an open source log database written in ANSI C language, supporting network, memory-based and persistent, Key-Value database, and provides API in multiple languages.

Unlike MySQL databases, Redis data is stored in memory. Its read and write speed is very fast and can handle more than 100000 read and write operations per second. Therefore, redis is widely used in caching. In addition, Redis is often used to do distributed locks. In addition, Redis supports transactions, persistence, LUA scripting, LRU-driven events, and multiple clustering scenarios.

Now that we know what redis is, let's talk about why Redis is so fast.

2. Why is redis so fast?

Let's explain one by one!

Implementation based on memory storage

Computer major students we all know that memory read and write is much faster than the disk, Redis is based on memory implementation of the database, compared with the data stored in the disk mysql and other databases, saving the consumption of disk Imano.

Efficient data structure

As we all know, in order to improve efficiency, mysql index chooses the data structure of B+ tree. For an application scenario, a reasonable data structure can make your application or program faster. Let's take a look at the data structure of Redis-the internal coding diagram:

String: dynamic string SDS

List: double-ended linked list LinkedList+ compressed linked list ziplist

Hash: compressed linked list ziplist+ dictionary hash table hashtable

Set: hashtable (+ inset)

Zset: compressed linked list ziplist+ hopping table skiplist

Let's talk about these internal codes:

1. SDS simple dynamic string

Let's compare it with char in C language.

String length processing: Redis gets the string length with a time complexity of O (1), while in C language, it needs to be traversed from scratch with a complexity of O (N).

Space pre-allocation: the more frequent string modification, the more frequent memory allocation, which will consume performance, while SDS modification and space expansion will allocate additional unused space and reduce performance loss.

Inert space release: when the SDS is shortened, instead of reclaiming the excess memory space, the free records the excess space, and if there are subsequent changes, directly use the space recorded in the free to reduce allocation.

Binary security: Redis can store some binary data. In C language, a string ends when it encounters'/ 0', while in SDS, it is the len attribute that marks the end of a string.

2. Dictionary

Redis as a KMel V-type in-memory database, all the keys are stored in a dictionary. A dictionary is a hash table, such as HashMap, and the corresponding value can be obtained directly through key. As for the characteristics of the hash table, the corresponding value can be obtained at O (1) time complexity.

3. Jump table

Jump table is a unique data structure of Redis, which is to increase multi-level index to improve search efficiency on the basis of linked list.

The jump table supports node lookup with average O (logN) and worst O (N) complexity, and can also be done through sequential operations.

Reasonable data coding

Redis supports multiple data types, each basic type, and possibly multiple data structures. When, what data structure and what encoding are used is the result of redis designers' summary and optimization.

String: if you store numbers, you use int type encoding; if you store non-numeric strings that are less than or equal to 39 bytes, if the embstr; is greater than 39 bytes, it is raw encoding.

List: if the number of elements in the list is less than 512, the value of each element in the list is less than 64 bytes (default), use ziplist encoding, otherwise use linkedlist encoding

Hash: if the number of hash type elements is less than 512, use ziplist encoding if all values are less than 64 bytes, otherwise use hashtable encoding.

Set: if all the elements in the collection are integers and the number of elements is less than 512, use intset encoding, otherwise use hashtable encoding.

Zset: when the number of elements in an ordered set is less than 128and the value of each element is less than 64 bytes, ziplist encoding is used, otherwise skiplist (jump table) encoding is used.

Reasonable threading model 1. Istroke O Multiplexing

The multiplexing technology of multi-channel epoll O enables a single thread to process multiple connection requests efficiently, while epoll is used as the implementation of the multiplexing technology. Moreover, Redis's own event handling model converts the connection, read and write, and shutdown in epoll into events, and does not waste too much time on the network Ibino.

2. What is Istroke O Multiplexing?

ICompo: network Ihamo

Multiplex: multiple network connections

Reuse: reuse the same thread.

IO multiplexing is actually a synchronous IO model that implements that a thread can monitor multiple file handles; once a file handle is ready, it can notify the application to read and write accordingly; and when no file handle is ready, it blocks the application and hands over the cpu.

3. Single thread model

Redis is a single-threaded model, which avoids the unnecessary context switching and competitive lock consumption of CPU. Also because it is single-threaded, if a command executes too long (such as the hgetall command), it will cause blocking. Redis is a database for fast execution scenarios. So be careful with commands such as smembers and lrange, hgetall, etc.

Redis 6.0introduces multithreading acceleration, which executes commands and operates on memory using a single thread.

Virtual memory mechanism

Redis directly built its own VM mechanism, unlike the general system will call system functions to deal with, will waste a certain amount of time to move and request.

What is the virtual memory mechanism of Redis?

The virtual memory mechanism is to temporarily swap infrequently accessed data (cold data) from memory to disk, thus freeing up valuable memory space for other data that needs to be accessed (hot data). The hot and cold data can be separated by the VM function, so that the hot data is still in memory and the cold data is saved to disk. In this way, the problem of slow access speed caused by insufficient memory can be avoided.

The above is all the contents of the article "sample Analysis of Redis efficiency reasons and data structures". Thank you for reading! Hope to share the content to help you, more related knowledge, welcome to follow the industry information channel!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.