Save 37% off PRO during our Black Friday Sale! »

Introduction to the Serverless paradigm

Introduction to the Serverless paradigm

Alex Casalboni and Austen Collins discuss the evolution of Serverless. Learn about the exciting new trend that's redefining the cloud computing industry in this in-depth webinar designed to teach you the basics of serverless computing and design.


Alex Casalboni

June 27, 2016


  1. Introduc)on  to  the   Serverless  Paradigm 6/27/2016

  2. About  us Alex  Casalboni Senior  So3ware  Engineer  & @alex_casalboni Austen

     Collins Founder  @ @austencollins AWS  Community  Hero Cloud  Evangelist  @  Cloud  Academy
  3. CharacterisHcs  of  Serverless  CompuHng Zero-­‐administra)on Pay-­‐per-­‐execu)on Event-­‐driven Microservice  oriented

    Func)on  as  the  unit  of  scale
  4. Myth  #1:  #NoServers There is always a server somewhere,

    just don't have to worry about it! FaaS  sounds  beDer? “there  are  no  servers  anymore” (Func)on  as  a  Service)
  5. Myth  #2:  #NoOps You will always need operations.
 Maybe not

    a DevOps team anymore.
 Devs should own their services. “opera4ons  are  not  required”
  6. Myth  #3:  #NoDevs 3rd party APIs are nothing new. Outsourcing

    makes development faster, but you can always BYOC. “you  just  don’t  write  any  code”
  7. Microservice  Principles Design  domain-­‐driven  services Always  hide  implementa)on  details Decentraliza)on

     is  key Achieve  failure  isola)on Automate  and  deploy  oNen  (CD/CI)
  8. AWS  Lambda Transparent  scalability,  elas)city  and  availability Developer-­‐friendly  opera)ons  (versions

     +  aliases) AWS   Lambda Na)ve  integra)ons  with  (almost)  everything  in  AWS Event-­‐driven  approach Never  pay  for  idle
  9. AWS  Lambda  +  API  Gateway  =  RESTful + Amazon  API

      Gateway AWS   Lambda RESTful  layer  (JSON/XML) Auth  layer  (SSL,  API  keys,  IAM  or  custom) Global  CDN  and  caching  (via  CloudFront) Logging,  throDling  &  staging  for  free Input/output  params  mapping  (VTL)
  10. AWS  Lambda  Use  Cases  (Event  Sources) Amazon  S3 Amazon

     Kinesis Amazon  DynamoDB AWS  CloudTrail Amazon  SNS Amazon  SES AWS  CloudForma)on Amazon  Cognito Amazon  SWF AWS  IoT AWS  CodePipeline Scheduled
  11. AWS  Lambda  LimitaHons Cold  start  issue Dependencies  management AWS  

    Lambda Deployment  package  size  &  local  storage Max  5  min  execu)on RAM,  CPU  &  Network  configura)on
  12. AWS  Lambda  RunHme  Environments Node.js  0.10 Node.js  4.3  (ES6)

    Python  2.7 Java  8
  13. AWS  Lambda  OrganizaHonal  Pa`erns 1.  Nanoservice  Pa`ern 1  func)on

     =  1  single  job least  amount  of  code  possible easiest  to  debug a  lot  of  Lambda  func)ons
  14. AWS  Lambda  OrganizaHonal  Pa`erns 2.  Microservice  Pa`ern 1  func)on

     =  a  few  related  jobs  (e.g.  CRUD) more  code  than  nanoservices reduced  number  of  func)ons reduced  “cold-­‐start”  impact
  15. AWS  Lambda  OrganizaHonal  Pa`erns 3.  New  Monolithic  Pa`ern only

     1  func)on  +  GraphQL least  amount  of  API  endpoints async  addi)onal  business  logic almost  no  “cold-­‐start” +
  16. The  Serverless  Ecosystem AWS  Lambda MicrosoN  Azure  Func)ons Google  Cloud

     Func)ons IBM  Bluemix  OpenWhisk Cloud  Plaaorms FuncHon  as  a  Service Auth0  (SSO) Misc Firebase  (DB) Algolia  (Search) Netlify  (CDN)
  17. Austen  Collins,  Founder 6/27/2016

  18. Thank  you  for  a`ending  :) Q  &  A 6/27/2016