Slide 1

Slide 1 text

Serverless the way microservices should be @ivanjov96

Slide 2

Slide 2 text

No content

Slide 3

Slide 3 text

How many of you are trying to scale your app?

Slide 4

Slide 4 text

How many DevOps engineers do you have?

Slide 5

Slide 5 text

How many do you really need? 0 Let’s see how

Slide 6

Slide 6 text

Ivan Jovanovic Senior software engineer @ nearForm https://ivanjov.com @ivanjov96

Slide 7

Slide 7 text

How did we get here? Microservices Monolith Serverless ?

Slide 8

Slide 8 text

Monolith

Slide 9

Slide 9 text

Monolith - Pros • Setup • One app to maintain • Deployment • Shared memory between functionalities • Monitoring, logging • Adding new functionalities

Slide 10

Slide 10 text

Monolith - Cons • Extending app without breaking other stuff • Slow delivery and deployments • Large codebase to maintain • Hard to understand • Slower development as it evolves

Slide 11

Slide 11 text

Microservices

Slide 12

Slide 12 text

Microservices - Pros • Service independence • Async communication between services • Independent deployment • Easier to understand, develop and test • Speeds up development • Enables continuous delivery and deployment

Slide 13

Slide 13 text

Microservices - Cons • Setup • Service discovery • Monitoring • Error tracking • Logging

Slide 14

Slide 14 text

How we do microservices now? ➕ ➕

Slide 15

Slide 15 text

Communication

Slide 16

Slide 16 text

How to fix this?

Slide 17

Slide 17 text

Serverless to the rescue!

Slide 18

Slide 18 text

What is serverless?

Slide 19

Slide 19 text

Serverless computing is a cloud- computing execution model in which the cloud provider dynamically manages the allocation of machine resources.

Slide 20

Slide 20 text

No content

Slide 21

Slide 21 text

No content

Slide 22

Slide 22 text

No content

Slide 23

Slide 23 text

Serverless - Pros • Setup, no provision needed • Simplifies packaging and deployment • Cost is small • Reduces the complexity • Automatic scaling

Slide 24

Slide 24 text

Serverless - Cons • Vendor lock in (but doesn’t need be) • No guarantee where will function be executed • Takes some time to handle first request - cold time

Slide 25

Slide 25 text

How does serverless “kill” DevOps? • No need to write any setup code, no containers, orchestration, nothing • Simple deployment • Functions are so small, no space for mistakes • It doesn’t “kill” DevOps, it just takes it to the next level

Slide 26

Slide 26 text

Who are the serverless providers? Amazon Lambda Google Cloud Functions Azure Functions

Slide 27

Slide 27 text

Comparison ⚔ ⚔

Slide 28

Slide 28 text

Comparison ❤ ❤

Slide 29

Slide 29 text

Serverless architecture

Slide 30

Slide 30 text

No content

Slide 31

Slide 31 text

and more…

Slide 32

Slide 32 text

But it costs

Slide 33

Slide 33 text

And more serverless functions…

Slide 34

Slide 34 text

For free With autoscaling

Slide 35

Slide 35 text

Serverless architecture Synchronous Asynchronous

Slide 36

Slide 36 text

Synchronous

Slide 37

Slide 37 text

Asynchronous

Slide 38

Slide 38 text

Logging AWS CloudWatch

Slide 39

Slide 39 text

Show me the code

Slide 40

Slide 40 text

No content

Slide 41

Slide 41 text

No content

Slide 42

Slide 42 text

No content

Slide 43

Slide 43 text

Tools

Slide 44

Slide 44 text

Claudia.js

Slide 45

Slide 45 text

No content

Slide 46

Slide 46 text

No content

Slide 47

Slide 47 text

Claudia API Builder

Slide 48

Slide 48 text

Serverless framework

Slide 49

Slide 49 text

No content

Slide 50

Slide 50 text

No content

Slide 51

Slide 51 text

No content

Slide 52

Slide 52 text

Who is using serveless?

Slide 53

Slide 53 text

Conclusion • Serverless, not just a buzzword • Pick provider and tools that you want • Free autoscaling • Have you infrastructure as a code • Fast development, deployment and delivery

Slide 54

Slide 54 text

Thank you! Blog ivanjov.com Twitter @ivanjov96