71 articles on Flink practice and principle analysis (necessary for interview)

De duplication series

One 、Flink Go to the first bullet :MapState duplicate removal

Two 、Flink De duplication second bullet ：SQL The way

3、 ... and 、Flink Go to the third bullet ：HyperLogLog duplicate removal

Four 、 About hyperloglog De optimize

5、 ... and 、 About bitmap Summary of the use of

6、 ... and 、Flink Go to the fourth bullet : bitmap Accurate weight removal

7、 ... and 、 Coding method implementation Split Distinct Aggregation function

Actual combat series catalog ：

One 、 Hotspot traffic optimization

Two 、 shunt

3、 ... and 、 Customize RetractStreamTableSink

Four 、 Customize UpsertStreamTableSink

5、 ... and 、Flink SQL Custom aggregate functions

6、 ... and 、Flink SQL in TableFunction Use analysis

7、 ... and 、StreamingFileSink Compress and merge small files

8、 ... and 、 The blood case caused by the change of parallelism

Window series directory ：

One 、window Implementation mechanism analysis

Two 、count window Decrypt

3、 ... and 、 Window utility trigger :ContinuousEventTimeTrigger

Dimension service series directory :

One 、 Dimension table service and Flink asynchronous IO
Two 、Mysql Dimension table Association ： Full load
3、 ... and 、Hbase Dimension table Association ：LRU Strategy
Four 、Redis Dimension table Association ： Real-time query
5、 ... and 、kafka Dimension table Association ： The way of broadcasting
6、 ... and 、 Custom asynchronous query

join series

One 、Flink DataStream in CoGroup Implementation principle and three join Realization

Two 、Flink intervalJoin Use and principle analysis

3、 ... and 、Flink join The terminator ：SQL Join

TopN series

One 、 window TopN Analysis and Implementation

Two 、 overall situation TopN Analysis and Implementation

The catalogue of Sanskrit series :

One 、 You should know Watermark

Two 、Flink Design and implementation of medium delay call

3、 ... and 、Flink SQL Decryption with the Revocable Mechanism in

Four 、Flink SQL Custom aggregate functions

5、 ... and 、Flink SQL in TableFunction Use analysis

6、 ... and 、2019 Year around Flink Some things done

7、 ... and 、 be based on Flink Log collection of

8、 ... and 、 About EventTime The problems brought about

Nine 、Flink Realization Locality Mode scheduling

Source analysis series

One 、StreamOperator Analysis of source code

Two 、 Customize StreamOperator

3、 ... and 、StreamTask Source code analysis

Four 、Flink Analysis of heartbeat mechanism

5、 ... and 、Flink be based on zookeeper High availability analysis

6、 ... and 、Flink RPC Implementation analysis

7、 ... and 、Flink Four elements flowing in

8、 ... and 、Flink A lock in

Nine 、Checkpoint Alignment mechanism source code analysis

Ten 、Flink in Watermark Timing generation source analysis

11、 ... and 、Flink: Two recursions completely understand operator chain

Task submission process analysis

One 、Flink1.10 Task submission and... Based on factory mode SPI Mechanism

Two 、Flink1.10 Task submission process analysis ( One )

3、 ... and 、Flink1.10 Task submission process analysis ( Two )

Time system series catalog ：

One 、 Overview of time system

Two 、Processing Time Source code analysis

3、 ... and 、Event Time Source code analysis

Four 、 Application analysis of time system in window function

5、 ... and 、ProcessFunction Use analysis

6、 ... and 、 Instance to explain ： How to do timing output

exactly-once Series catalog ：

One 、 Two stage submission Overview

Two 、 Two phase submission implementation analysis

3、 ... and 、StreamingFileSink analysis

Four 、 Transactional output implementation

5、 ... and 、 The final consistent implementation

Operation and maintenance series catalog ：

One 、 Task log collection

Two 、 Customize metric Monitor inflow 、 Output

3、 ... and 、 Remember a Flink write in Kafka Pit point

Four 、Flink per-Job Pattern InfluxdbReporter Report JobName

CEP series

One 、 One Flink-Cep Use cases

Two 、Flink-Cep Realize dynamic update of rules

Clickhouse series

One 、Clickhouse LB practice

Two 、 Practice from the demand scenario Clickhouse