Ali's redis interview question is too difficult, isn't it? I was pressed on the ground and rubbed

Preface

As the baldhead of this time code , Yes Redis It must be no stranger , If even Redis It's not used , I'm sorry to go out for an interview , I'm not sure how many times I've been beaten by the interviewer . After all, now Internet companies and some start-ups have to use Redis, Like Amazon 、 Google 、 Ali 、 Tencent should use , We can see that we are proficient in Redis It's really necessary to use . For myself, I hung up on the interview with ALI last year redis On the issue of , I'm not familiar with myself redis The second is that Ali's interview question is too difficult , Just give me a few words , See if you can answer ：

• redis Why key,value Of , Why not support SQL Of ？
• redis Multi thread or single thread ？（ Answer the single thread, please go back , Why do you come back , Please look down ）
• redis The persistence of RDB and AOF How does the restart service load ？(10 personal 9 A wrong answer )
• redis How to plan a cluster ？AKF/CAP How to implement and Design ？
• 10 Ten thousand users a year 365 How to use the login status of days redis Storage , And quickly retrieve the active users in any time window ？
• redis Of 5 Kind of Value Type you've used several , Can you give me an example ？
• 100 Million concurrent 4G data ,10 Million concurrent 400G data , How to design Redis storage ？

As excellent as I was, I was pressed hard on the ground and rubbed , In order to solve this problem, I specially got a copy from Tencent cloud's good brother, which was coded by Tencent cloud God himself “redis In depth notes ”, Let's share it for free today , There is no nonsense in the notes , The whole article looks at the essence ！

Due to space limitations, only screenshots can be shown below , Need full version pdf My friend, please like, pay attention and support , Follow me and send a private message “redis” You can get this information for free

PART1：Redis In depth notes begin

1.Redis What can you do with it ？

• from Redis What I think of in the interview
• Redis What can be done ？

2.Redis Infrastructure

• Redis install
• Redis Basic data structure
• General rules of container data structure
• About Redis Some thoughts on using

PART2：Redis Summary of the application of

1. Distributed lock

• Distributed lock
• Timeout problem
• Reentrancy

2. Delay queue

• Asynchronous message queuing ？
• What to do if the queue is empty ？
• Queue delay
• The idle connection is automatically disconnected
• Lock conflict handling
• The implementation of delay queue
• Further optimization

3. Bitmap

• Basic use
• Statistics and search
• Magic command bitfield

4.HyperLogLog

• Usage method
• pfadd This pf What does that mean? ？
• pfmerge Suitable for what field ？
• matters needing attention
• HyperLogLog Realization principle
• pf Why is the memory usage of 12k？

5. The bloon filter

• What is a bloon filter ？
• Redis Bloom filter in
• Basic use of Bloom filter
• matters needing attention
• The principle of Bloom filter
• Space occupancy estimates
• When the actual element exceeds , How the miscalculation rate will change ？
• It will not be on Redis4.0 What do I do ？
• Other applications of Blum filter

6. Simple current limiting

• How to use Redis To implement a simple current limiting strategy ？

7. Funnel restriction

• Redis-Cell
• Some thoughts

8.GeoHash

• Use the database to count people in the neighborhood
• GeoHash Algorithm
• Redis Of Geo The command basically uses

9.Scan

• scan Based on using
• Structure of dictionary
• scan traversal order
• Dictionary expansion
• Compare the traversal order before and after capacity expansion
• Progressive type rehash
• added scan Instructions
• Big Key scanning

PART3：Redis Principle

1. Threads IO Model

• Non blocking IO
• Event polling （ Multiplexing ）
• Command queue
• Response queue
• Timing task

2. Communication protocol

• RESP（Redis Serialization Protocol）
• client -> Server side
• Server side -> client

3. Persistence

• Snapshot principle
• fork（ Multi process ）
• AOF principle
• AOF rewrite
• fsync
• Operation and maintenance
• Redis4.0 Mix persistence

Due to space limitations, only screenshots can show this information , Need full version pdf My friend, please like, pay attention and support , Follow me and send a private message “redis” You can get this information for free

4. The Conduit

• Redis Message interaction
• Pipeline pressure test
• In depth understanding of the nature of the pipeline

5. Business

• Redis Basic use of transactions
• Atomicity
• discard（ discarded ）
• Optimize
• Watch

6.PubSub

• Message multicast
• PubSub
• Mode subscription
• Message structure
• PubSub shortcoming

Due to space limitations, only screenshots can be shown below , Need full version pdf My friend, please like, pay attention and support , I put the free access method in the comment area

7. Small object compression

• 32bit vs 64bit
• Small object compression storage （ziplist）
• Memory recovery mechanism
• Memory allocation algorithm

8. Master slave synchronization

• CAP principle
• Final agreement
• Master slave synchronization
• The incremental synchronization
• Snapshot synchronization
• Add slave
• Copy without disk
• Wait Instructions

PART4：Redis colony

1.Sentinel

• Lost message
• Sentinel Basic use

2.Codis

• Codis Fragmentation principle
• Different Codis How to synchronize the slot relationship between instances ？
• Capacity expansion
• Automatic equalization
• Codis The price of
• Codis The advantages of
• MGET The operation of the command
• Structural change
• Codis Embarrassment
• Codis Back office management

3.Cluster

• Slot location algorithm
• Jump
• transfer
• Fault tolerance
• The network jitter
• Maybe offline （PFAIL-Possibly Fail） And confirm offline （Fail）
• Cluster Basic use
• Slot migration perception
• Cluster change awareness

PART5：Redis expand

1.Stream

• news ID
• The message content
• Additions and deletions
• Independent consumption
• Create a consumption group
• consumption
• Stream What to do if there is too much news ？
• If the news is forgotten ACK What will happen ？
• PEL How to avoid message loss ？
• Stream High availability
• Partition Partition

2.Info Instructions

• Redis How many instructions are executed per second ？
• Redis How many clients are connected ？
• Redisn How much memory is used ？
• How large is the copy backlog cache ？

Due to space limitations, only screenshots can show this information , Need full version pdf My friend, please like, pay attention and support , Follow me and send a private message “redis” You can get this information for free

3. Let's talk about distributed locks

• Redlock Algorithm
• Redlock Use scenarios

4. Expiration strategy

• overdue key aggregate
• Timing scan strategy
• Expiration policy of slave Library

5.LRU

• LRU Algorithm
• The approximate LRU Algorithm

6. Lazy delete

• Redis Why lazy delete （lazy free）？
• flush
• Asynchronous queues
• AOF Sync Very slow, too
• More asynchronous delete points

7. Use... Gracefully Jedis

• retry

8. Protect Redis

• Command security
• Port security
• Lua Script security
• SSL agent

9.Redis Secure communications

• spiped principle
• spiped Getting started

PART6： Source code

1. Explore the internal structure of strings

• embstr vs raw
• Expansion strategy

2. Explore the inside of the dictionary

• dict internal structure
• Progressive type rehash
• Lookup process
• hash function
• hash attack
• Capacity expansion condition
• Reduction conditions
• set Structure

3. Explore the inside of the compressed list

• Add elements
• update cascade
• IntSet Set of small integers

4. Explore the quick list inside

• Every ziplist How many elements are there ？
• Compression depth

5. Explore the internal structure of the jump list

• The basic structure
• Lookup process
• Random number of layers
• Insertion process
• Delete process
• The update process
• If score It's all the same ？
• How is the element ranking calculated ？

6. Explore the inside of the compact list

• update cascade
• replace ziplist

summary

Last , What I want to say is , Learning is not difficult , And it's about persistence , Especially after we got involved in the work , It's even harder to keep learning . But for programmers , Learning is the foundation of a career , If you give up learning , It's a matter of time before we are eliminated by the market . therefore , Learning more new knowledge is a bigger investment for yourself .

above , To this end .

Due to space limitations, only part of the content can be displayed. If you need this information, please praise, pay attention and support , Follow me and send a private message “redis” You can get this information for free