Predictive Maintenance Pipeline using Kafka Connect, Streams and KSQL

Slide 1

Slide 1 text

KSQL and Kafka Streams for Predictive Maintenance Pipeline

Slide 2

Slide 2 text

data engineer at

Slide 3

Slide 3 text

Slide 4

Slide 4 text

started looking at Kafka in April

Slide 5

Slide 5 text

Slide 6

Slide 6 text

Predictive Maintenance

Slide 7

Slide 7 text

No content

Slide 8

Slide 8 text

No content

Slide 9

Slide 9 text

No content

Slide 10

Slide 10 text

Component A - Health Last serviced: 1 October 2018 30% B77W 9V-123

Slide 11

Slide 11 text

aircraft sensors file system source connector ksql kafka streams app ML model web app

Slide 12

Slide 12 text

No content

Slide 13

Slide 13 text

aircraft sensors file system source connector ksql kafka streams app ML model web app 1 2

Slide 14

Slide 14 text

Source Connector custom connector written in Java

Slide 15

Slide 15 text

KSQL filtering based on plane types, tail number, etc.

Slide 16

Slide 16 text

Kafka Streams basic transformation, calls Machine Learning job

Slide 17

Slide 17 text

Why Kafka?

Slide 18

Slide 18 text

decoupling

Slide 19

Slide 19 text

aircraft sensors file system source connector ksql kafka streams app ML model web app Java Scala C# Python

Slide 20

Slide 20 text

decoupling

Slide 21

Slide 21 text

aircraft sensors file system source connector ksql kafka streams app ML model web app ksql kafka streams app ML model web app ksql kafka streams app ML model web app one pipeline for each aircraft component

Slide 22

Slide 22 text

Why Kafka?

Slide 23

Slide 23 text

offset management

Slide 24

Slide 24 text

Why Kafka Streams?

Slide 25

Slide 25 text

fault-tolerant

Slide 26

Slide 26 text

No content

Slide 27

Slide 27 text

Why Kafka Streams?

Slide 28

Slide 28 text

scalability

Slide 29

Slide 29 text

aircraft sensors file system source connector ksql kafka streams app ML model web app 1 2

Slide 30

Slide 30 text

aircraft sensors file system source connector ksql kafka streams app ML model web app 1 2 kafka streams app kafka streams app

Slide 31

Slide 31 text

Source: https://kafka.apache.org/0102/documentation/streams/architecture#streams_architecture_threads

Slide 32

Slide 32 text

Why Kafka Streams?

Slide 33

Slide 33 text

easy-to-use APIs

Slide 34

Slide 34 text

val builder: StreamsBuilder = new StreamsBuilder val inputLines: KStream[String, String] = builder.stream[String, String](inputTopic) val outputLines: KStream[String, SomeObject] = inputLines.map[String, SomeObject]((_: String,value: String) => { // perform some ETL (s"$filename", // build resulting object) }) outputLines.to(s"$outputTopic") val streams: KafkaStreams = new KafkaStreams(builder.build(), config) streams.start()

Slide 35

Slide 35 text

create stream 2018_sensor_stream with (kafka_topic=‘2018_sensor_stream’, value_format='avro') as select plane, tail, flight, year, month, date from sensor_stream where year >= 2018;

Slide 36

Slide 36 text

Learning

Slide 37

Slide 37 text

No content

Slide 38

Slide 38 text

No content

Slide 39

Slide 39 text

Summary

Slide 40

Slide 40 text

Predictive Maintenance • Using Machine Learning model to determine health of critical aircraft components.

Slide 41

Slide 41 text

Why Kafka? • Decoupled architecture. • Offset management.

Slide 42

Slide 42 text

Why Kafka Streams and KSQL? • Fault tolerant. • Scalable • Easy-to-use APIs.