I Love APIs 2015 Chris Munns, Amazon @chrismunns http://www.amazon.com/ As computing costs decreased and computing power grew over time, so increased the complexity of the problems computers were called to solve and complexity of software. Enterprise applications quickly went through the stage of monolithic applications to client-server to multiple tier and beyond – to the land of massively distributed architectures. We arrived at the point where enterprise software is well beyond the capability of a single person or even a reasonably practical group of people to understand and control. Are microsevices the answer? Join Chris Munns to learn about how microservices are scaled at Amazon.

1. Microservices at Amazon
Chris Munns
Business Development Manager - DevOps

2. About me:
Chris Munns - munns@amazon.com, @chrismunns
– Business Development Manager – DevOps
– New Yorker
– Previously:
• AWS Solutions Architect 2011-2014
• Lead of Infrastructure/DevOps @hingeapp
• Formerly on operations teams @Etsy and @Meetup
• Little time at a hedge fund, Xerox and others
– Rochester Institute of Technology: Applied Networking and Systems Administration
’05
– Internet infrastructure geek

3. The Micro-elephant in the room
https://www.flickr.com/photos/davidrosenphotography/14711027415/

4. Microservices
“is a software architecture style in which complex
applications are composed of small, independent
processes communicating with each other using
language-agnostic APIs. These services are
small, highly decoupled and focus on doing a
small task, facilitating a modular approach to
system-building.” - Wikipedia
https://en.wikipedia.org/wiki/Microservices

5. Microservices
Smart people are talking about it:
• Sam Newman - book “Building
Microservices” O’Reilly Publishing
• Adrian Cockcroft – numerous talks,
presentations, blog posts
“Loosely coupled service oriented
architecture with bounded contexts”
• Martin Fowler – numerous blog posts
• 2 sessions at AWS Re:Invent 2014
• ~8 sessions at AWS Re:Invent 2015

6. But is it just a fad?
Phil Calcado(ex-SoundCloud) in his Sept 8th post entitled “How we ended up
with microservices.” highlighted the recent growth of the term:
Data from Google Trends
http://philcalcado.com/2015/09/08/how_we_ended_up_with_microservices.html

7. But is it just a fad?
As they say in dramatic police dramas….Let’s ENHANCE!:
Data from Google Trends

8. Lots of folks are doing it
• Gilt: “From Monolith Ruby App to Distributed Scala Micro-
Services” NYC Tech Talks
• Nike: “Nike’s Journey to Microservices” – AWS Re:Invent
2014
• SoundCloud: ”Building Products at SoundCloud—Part III:
Microservices in Scala and Finagle”
• Capital One: “Lack Of Legacy Lets Capital One Build
Nimble Infrastructure” – ThePlatform.net
• Hailo: “A Journey into Microservices” – sudo.hailoapp.com

9. Brilliant folks are sharing tools for it
• Netflix: Dozens of tools - https://netflix.github.io/
• Airbnb: SmartStack - http://nerds.airbnb.com/
smartstack-service-discovery-cloud/
• Twitter: Mesos - https://mesos.apache.org/
• LinkedIn: Rest.li - https://github.com/linkedin/
rest.li

11. Monolith development lifecycle
developers
releasetestbuild
delivery pipelineapp

12. • Single-purpose
• Connect only
through APIs
• Connect over HTTPS
• Largely “black
boxes” to each other
• “Microservices”

13. Microservices vs. SOA
Wikipedia: “distinct from a Service-oriented architecture (SOA)
in that the latter aims at integrating various (business)
applications whereas several microservices belong to one
application only”
Martin Fowler: “This common manifestation of SOA has led
some microservice advocates to reject the SOA label entirely,
although others consider microservices to be one form of SOA”
Oracle: “The picture that emerges is not of microservices as an
alternative to SOA, but rather as a way to restore flexibility that
may have been lost in SOAs that became too rigid and
monolithic.” https://en.wikipedia.org/wiki/Microservices
http://martinfowler.com/articles/microservices.html#MicroservicesAndSoa
http://www.oracle.com/technetwork/issue-archive/2015/15-mar/o25architect-2458702.html

14. Microservices:
• Many very small components
• Business logic lives inside of
single service domain
• Simple wire protocols(HTTP
with XML/JSON)
• API driven with SDKs/Clients
SOA:
• Fewer more sophisticated
components
• Business logic can live across
domains
• Enterprise Service Bus like
layers between services
• Middleware
Microservices vs. SOA
14

15. • Two-pizza teams
• Full ownership
• Full accountability
• Aligned incentives
• “DevOps”

16. Wait, now we’re
talking about
Pizza?
16
©2015 Apigee. All Rights Reserved.
Image By: Chris Munns – munns@amazon.com

17. How do Two Pizza Teams work?
We call them “Service teams”
• Own the “primitives” they build:
– Product planning (roadmap)
– Development work
– Operational/Client support work
• “You build it, you run it”
• Part of a larger concentrated org (Amazon.com,
AWS, Prime, etc)
17

18. Who Does QA?
18
The Two
Pizza
Team

19. Who Does
On Call?
19
Image By: Chris Munns – munns@amazon.com
The Two
Pizza
Team

20. What does Ops
Do?
20
Not
Exist

21. What about Ops/QA/Etc?
Everyone exists on a “service team” focused on their
primitive(s):
• SDE’s focused on developing
• PM’s focused on product direction
• TPM’s help drive development
• SE’s focused on infra/tooling
• SDET’s focused on test excellence throughout the
organization
Some folks are shared across the org, some on individual
teams
21

22. Boy, that sounds like a lot of freedom?
It is! Teams are empowered and also held to high
standards:
• Thorough onboarding/training
• Patterns/practices defined at scale and with 20+
years of organizational knowledge
• Regular technical and business metric reviews
• Regular sharing of new tools, services, technologies,
etc, by internal subject matter experts
22

23. Missing tools
developers delivery pipelineservices
???

24. • Self-service
• Technology-agnostic
• Encourage best
practices
• Single-purpose
services

25. • Deployment service
• No downtime
deployments
• Health checking
• Versioned artifacts
and rollbacks

26. • Continuous delivery
• From check-in to
production
• Automated actions
and transitions
• >90% of teams
Pipelines

27. Microservice development lifecycle
developers delivery pipelinesservices
releasetestbuild
releasetestbuild
releasetestbuild
releasetestbuild
releasetestbuild
releasetestbuild

28. = 50 million deployments a year
Thousands of teams
× Microservice architecture
× Continuous delivery
× Multiple environments

29. I’m sold, how do
I make my own
pizza?
29
©2015 Apigee. All Rights Reserved.
Image By: Chris Munns – munns@amazon.com

30. How to make pizza?
• Yeast
• Water
• Sugar
• Olive oil
• Tomato sauce
• Cheese
• <insert favorite toppings here>
30

31. No Chris, I mean in
the organizational/
technological sense
31
©2015 Apigee. All Rights Reserved.
Image By: Chris Munns – munns@amazon.com

32. The keys to your own Two Pizza teams & Microservices
Culture
Practices
Tools
32

33. The keys to your own Two Pizza teams & Microservices
Culture
– Ownership & Accountability go hand in hand
– Larger teams typically move slower than smaller teams
– Insist on standards of excellence, but be flexible on the how
Practices
– Continuous Integration & Delivery
– Simplify operational tasks
Tools
– CI & CD tools
– Infrastructure management
– Metrics/Monitoring/Logging
– Communication and Collaboration
33

34. The keys to your own Two Pizza teams & Microservices
Culture
– Ownership & Accountability go hand in hand
– Larger teams typically move slower than smaller teams
– Insist on standards of excellence, but be flexible on the how
Practices
– Continuous Integration & Delivery
– Simplify operational tasks
Tools
– CI & CD tools
– Infrastructure management
– Metrics/Monitoring/Logging
– Communication and Collaboration
34

35. The keys to your own Two Pizza teams & Microservices
Culture
– Ownership & Accountability go hand in hand
– Larger teams typically move slower than smaller teams
– Insist on standards of excellence, but be flexible on the how
Practices
– Continuous Integration & Delivery
– Simplify operational tasks
Tools
– CI & CD tools
– Infrastructure management
– Metrics/Monitoring/Logging
– Communication and Collaboration
35

36. The keys to your own Two Pizza teams & Microservices
Culture
– Ownership & Accountability go hand in hand
– Larger teams typically move slower than smaller teams
– Insist on standards of excellence, but be flexible on the how
Practices
– Continuous Integration & Delivery
– Simplify operational tasks
Tools
– CI & CD tools
– Infrastructure management
– Metrics/Monitoring/Logging
– Communication and Collaboration
36
Internal “Service teams”
should own, support,
train, drive excellence
of these tools

37. Apollo and Pipelines as AWS Services:
CodePipeline
launched July 2015
CodeDeploy
launched Nov 2014

38. Establishing a pattern for services and clients
It’s important that the organization isn’t reinventing the
wheel on every new service:
• How are clients going to communicate?
• What cross service authorization requirements are there?
• How do services prevent abuse?
• How do you quickly build clients against a service?
• How do services handle discovery of others services and resources?

39. Use an API Gateway!
Internet
Mobile Apps
Websites
Services
API
Gateway
AWS Lambda
functions
API Gateway
Cache
Endpoints on
Amazon EC2
Any other publicly
accessible endpoint
Amazon
CloudWatch
Monitoring
Amazon
CloudFront

40. HOST
METRICS
SERVICE
METRICS
LOG
ANALYSIS
BUILD
METRICS

41. FIN, ACK
This is a big topic that is hard to cover in a single session, day,
week! Key pointers:
• Microservices patterns are changing how we are building APIs
and our applications in general
• Team structure plays an incredible role in building, running,
scaling Microservices
• Focusing on flexible tooling that provides standards can help an
organization focus on what matters
• Enable the smart people you hire to do great things by giving
them both ownership and accountability on those things
success
41

42. Thank you
42