The State of Serverless

Slide 1

Slide 1 text

the state of serverless

Slide 2

Slide 2 text

Maciej Winnicki Principal Software Engineer Serverless, Inc.

Slide 3

Slide 3 text

@mthenw

Slide 4

Slide 4 text

1973

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

https://blog.prismalytics.io/serverless-frenzy-a29fefc1514f |

Slide 11

Slide 11 text

“ The next day saw an unforgettable orgy of one-liners as everybody joined in the excitement of plumbing http://www.cs.dartmouth.edu/~doug/reader.pdf

Slide 12

Slide 12 text

tail -f syslog | grep error | tee errors

Slide 13

Slide 13 text

No content

Slide 14

Slide 14 text

unix philosophy

Slide 15

Slide 15 text

“ Doug McIlroy, E. N. Pinson, B. A. Tague (8 July 1978). "Unix Time-Sharing System: Foreword". pp. 1902–1903. Design and build software, even operating systems, to be tried early, ideally within weeks. Don't hesitate to throw away the clumsy parts and rebuild them.

Slide 16

Slide 16 text

MAKE PROGRAMMERS MORE PRODUCTIVE

Slide 17

Slide 17 text

https://medium.com/apex-serverless/introducing-apex-800824ﬀaa70

Slide 18

Slide 18 text

what is serverless

Slide 19

Slide 19 text

(re)evolution of the cloud

Slide 20

Slide 20 text

focus on business logic less concerns about stack implementation https://medium.com/openwhisk/five-minute-intro-to-open-source-serverless-development-with-openwhisk-328b0ebfa160 bare metal IaaS PaaS functions hardware abstraction OS abstraction runtime abstraction

Slide 21

Slide 21 text

Function-as-a-Service

Slide 22

Slide 22 text

Function-as-a-Service Backend-as-a-Service + https://martinfowler.com/articles/serverless.html

Slide 23

Slide 23 text

no server/container/process management http://blog.rowanudell.com/the-serverless-compute-manifesto/ costs scale with usage / never pay for idle auto-scale / auto-provision

Slide 24

Slide 24 text

global availability

Slide 25

Slide 25 text

Function-as-a-Service is event-driven

Slide 26

Slide 26 text

use cases

Slide 27

Slide 27 text

back-end services / web apps / IoT https://www.youtube.com/watch?v=b1t78P_1FT4 data processing infrastructure automation and more…

Slide 28

Slide 28 text

challenges

Slide 29

Slide 29 text

functions are like microservices but smaller

Slide 30

Slide 30 text

monitoring & logging debugging & diagnostics local development

Slide 31

Slide 31 text

vendor lock-in

Slide 32

Slide 32 text

latency / cold start

Slide 33

Slide 33 text

providers

Slide 34

Slide 34 text

No content

Slide 35

Slide 35 text

No content

Slide 36

Slide 36 text

No content

Slide 37

Slide 37 text

• Node.js • Java • Python • C# Runtimes

Slide 38

Slide 38 text

• Amazon S3 • Amazon DynamoDB • Amazon Kinesis Streams • Amazon Simple Notification Service • Amazon Simple Email Service • Amazon Cognito • AWS CloudFormation • Amazon CloudWatch Logs • Amazon CloudWatch Events • AWS CodeCommit • Scheduled Events (powered by Amazon CloudWatch Events) • AWS Config • Amazon Echo • Amazon Lex Amazon • API Gateway Events http://docs.aws.amazon.com/lambda/latest/dg/invoking-lambda-function.html#supported-event-source-cloudformation

Slide 39

Slide 39 text

Monitoring & Logging • Logs and metrics pushed to CloudWatch • Available metrics: • Invocations • Errors • Duration • Throttles

Slide 40

Slide 40 text

Debugging & Diagnostics • X-Ray - distributed tracing system • Support for: • Java • Node.js

Slide 41

Slide 41 text

X-Ray Demo

Slide 42

Slide 42 text

Local Development • no oﬀicial tools

Slide 43

Slide 43 text

Local Development

Slide 44

Slide 44 text

Local Development • API Gateway • Kinesis • DynamoDB • DynamoDB Streams • Elasticsearch • S3 • Firehose • Lambda • SNS • SQS • Redshift • ES (Elasticsearch Service) • SES • Route53 • CloudFormation • CloudWatch

Slide 45

Slide 45 text

Ecosystem • Step Functions • workflows • coordinate functions

Slide 46

Slide 46 text

No content

Slide 47

Slide 47 text

Summary: AWS is a leader in serverless.

Slide 48

Slide 48 text

No content

Slide 49

Slide 49 text

Google Cloud Functions

Slide 50

Slide 50 text

in beta

Slide 51

Slide 51 text

Runtimes • Node.js

Slide 52

Slide 52 text

Events • HTTP request • Cloud Pub/Sub • Cloud Storage

Slide 53

Slide 53 text

Monitoring & Logging • Logs and metrics pushed to Stackdriver Logging • Errors captured by Stackdriver Error Reporting • Available metrics: • execution time • execution count • memory usage

Slide 54

Slide 54 text

• Debugging via Stackdriver Debugger Debugging & Diagnostics

Slide 55

Slide 55 text

Stackdriver Debugger Demo

Slide 56

Slide 56 text

• Cloud Functions Local Emulator Local Development

Slide 57

Slide 57 text

• Cloud Functions For Firebase • Events: • Realtime Database Triggers • Firebase Authentication Triggers • Google Analytics for Firebase Triggers Ecosystem

Slide 58

Slide 58 text

Summary: GCP oﬀers very basic serverless features.

Slide 59

Slide 59 text

No content

Slide 60

Slide 60 text

Runtimes • Node.js • C# • F# • Python • PHP

Slide 61

Slide 61 text

Events • HTTP request • Schedule • Azure Storage blob storage and queues • Azure Event Hubs events • Azure Service Bus queues and topics

Slide 62

Slide 62 text

Monitoring & Logging • Logs and metrics pushed to Application Insights • Metrics available per server: • response time • requests • failed requests

Slide 63

Slide 63 text

Debugging & Diagnostics • Debugging via local development (Visual Studio)

Slide 64

Slide 64 text

Local Development • Azure Functions Core Tools

Slide 65

Slide 65 text

Ecosystem • Logic Apps • business processes • workflows • integration with SaaS products

Slide 66

Slide 66 text

Summary: Decent serverless oﬀering. Logic Apps look promising.

Slide 67

Slide 67 text

No content

Slide 68

Slide 68 text

Runtimes • Node.js • Swift • Python • Java • anything via Docker

Slide 69

Slide 69 text

Events • HTTP request • Schedule (Alarms) • GitHub events • IBM Cloudant DB changes • IBM Message Hub events • IBM Push Notifications events • custom events via triggers

Slide 70

Slide 70 text

Monitoring & Logging • Logs available via dashboard and CLI • Metrics: • number of invocations • number of failed invocations • completion time

Slide 71

Slide 71 text

Debugging & Diagnostics • no oﬀicial tools

Slide 72

Slide 72 text

Local Development • no oﬀicial tools

Slide 73

Slide 73 text

Ecosystem

Slide 74

Slide 74 text

Summary: Open-source FaaS project built with extensibility in mind.

Slide 75

Slide 75 text

other projects

Slide 76

Slide 76 text

Kubeless

Slide 77

Slide 77 text

Fission

Slide 78

Slide 78 text

Funktion

Slide 79

Slide 79 text

FaaS

Slide 80

Slide 80 text

No content

Slide 81

Slide 81 text

FaunaDB • database from the team that scaled Twitter • global replication/global consistency • supported data models: • relational • document • object-oriented • graph

Slide 82

Slide 82 text

the company

Slide 83

Slide 83 text

No content

Slide 84

Slide 84 text

A CLI tool that manages the lifecycle of you serverless application.

Slide 85

Slide 85 text

Features • application lifecycle management • manage code & infrastructure • scaﬀolding & automation • multi-provider (AWS, GCloud, Azure, IBM OpenWhisk) • local development • extensibility • community