Slide 1

Slide 1 text

DISTRIBUTED SYSTEM TRACING @dustyburwell Monitoring Performance in Distributed Systems

Slide 2

Slide 2 text

No content

Slide 3

Slide 3 text

Dusty Burwell @dustyburwell

Slide 4

Slide 4 text

No content

Slide 5

Slide 5 text

No content

Slide 6

Slide 6 text

No content

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

Traditional Options

Slide 11

Slide 11 text

CPU Tracing Statistical or Detailed

Slide 12

Slide 12 text

No content

Slide 13

Slide 13 text

No content

Slide 14

Slide 14 text

No content

Slide 15

Slide 15 text

Memory Dumping

Slide 16

Slide 16 text

Structured Log Aggregation

Slide 17

Slide 17 text

Structured Log Aggregation a.k.a. read the f’n error

Slide 18

Slide 18 text

Microservices Tracing

Slide 19

Slide 19 text

Dapper http://bit.ly/cm-dapper

Slide 20

Slide 20 text

Vocabulary

Slide 21

Slide 21 text

Trace

Slide 22

Slide 22 text

No content

Slide 23

Slide 23 text

Trace ID

Slide 24

Slide 24 text

Span

Slide 25

Slide 25 text

Trace ID

Slide 26

Slide 26 text

Trace ID

Slide 27

Slide 27 text

Trace ID Span ID

Slide 28

Slide 28 text

Trace ID Span ID

Slide 29

Slide 29 text

Trace ID Span ID Parent Span ID

Slide 30

Slide 30 text

Trace ID Span ID Parent Span ID Client Start - timestamp Server Start - timestamp Server End - timestamp Client End - timestamp

Slide 31

Slide 31 text

Annotation

Slide 32

Slide 32 text

Trace ID Span ID Parent Span ID Client Start - timestamp Server Start - timestamp Server End - timestamp Client End - timestamp

Slide 33

Slide 33 text

Trace ID Span ID Parent Span ID Client Start - timestamp Server Start - timestamp Server End - timestamp Client End - timestamp

Slide 34

Slide 34 text

Trace ID Span ID Parent Span ID Client Start - timestamp Server Start - timestamp Server End - timestamp Client End - timestamp Service Name Endpoint Is Error? Error code Etc.

Slide 35

Slide 35 text

No content

Slide 36

Slide 36 text

Parts

Slide 37

Slide 37 text

Instrumentation

Slide 38

Slide 38 text

Transport libraries (HTTP, etc) Middleware or similar Headers passed from instrumented libraries and received by middleware pass along Trace ID and Parent Span ID

Slide 39

Slide 39 text

Sampling

Slide 40

Slide 40 text

Tryfer https://github.com/tryfer

Slide 41

Slide 41 text

Data Pipeline

Slide 42

Slide 42 text

Aggregation, Visualization, and Reporting

Slide 43

Slide 43 text

Specialized Tooling

Slide 44

Slide 44 text

No content

Slide 45

Slide 45 text

Generalized Tooling

Slide 46

Slide 46 text

Generalized Tooling e.g. MapReduce

Slide 47

Slide 47 text

Challenges

Slide 48

Slide 48 text

Achieving Buy In

Slide 49

Slide 49 text

Achieving Buy In a.k.a. Chicken and Egg

Slide 50

Slide 50 text

Polyglot Probs

Slide 51

Slide 51 text

No content

Slide 52

Slide 52 text

Callback Oriented Frameworks

Slide 53

Slide 53 text

Exercise for the Reader

Slide 54

Slide 54 text

Dapper http://bit.ly/cm-dapper

Slide 55

Slide 55 text

Zipkin http://zipkin.io/

Slide 56

Slide 56 text

Working Groups Distributed Tracing and SOA

Slide 57

Slide 57 text

Log Aggregation Splunk, ELK, etc.

Slide 58

Slide 58 text

NewRelic http://newrelic.com/

Slide 59

Slide 59 text

Stackify http://stackify.com/