This talk introduces microservices as a tool in an API developer's arsenal. We'll introduce what they are, see how and why they could fit into a modern application (and when they may not), and tools that will make dealing with a microservices architecture easier than ever before.

1. M I C R O S E R V I C E S
D E C O U P L I N G A P I S T H R O U G H
David Simons
@SwamWithTurtles

2. M I C R O S E R V I C E S
D E C O U P L I N G A P I S T H R O U G H
David Simons
@SwamWithTurtles

3. W H O A M I ?
• David Simons
(@SwamWithTurtles)
• Freelance Tech Lead/
Consultant
• Hacker specialising in
Node and Java

4. M Y E X P E R I E N C E
W I T H A P I S
• Focused on web
development
• Utilise data from our own,
or third party APIs to affect
rendered data

5. W H AT I WA N T T O
TA L K A B O U T…
• What problems do
microservices solve?
• What is a microservice?
• An example!
• When are microservices
wrong?
• How can I start
implementing
microservices?

6. W H AT P R O B L E M S D O
M I C R O S E R V I C E S S O LV E ?
Q U E S T I O N 1 :

7. W H AT D O W E
A S P I R E F O R I N
A G O O D A P I ?

8. W H AT M A K E S A G O O D A P I ?
HATEOS-y
Sensibly named
Ignorant of Consumers
Well Maintained
Tested
Appropriate
Extensible
Hard to Misuse
Secure Useful
High Quality Data
RESTful
Easy to Adopt
Resilient
Performant
Versioned
Declarative
Flexible output formats
Stable
Well Documented

9. D E C O U P L E D
A G O O D A P I I S …

10. R E U S A B L E
A G O O D A P I I S …

11. S C A L A B L E
A G O O D A P I I S …

12. A M I C R O S E R V I C E
A R C H I T E C T U R E M A K E S
Y O U R A P I S M O R E :
D E C O U P L E D , R E U S A B L E
A N D S C A L A B L E
M Y C L A I M …

13. W H AT I S A
M I C R O S E R V I C E ?
Q U E S T I O N 2 :

14. Independently deployable services
focused around business concerns,
which contain all your logic.

15. F O C U S E D
A R O U N D
B U S I N E S S
C O N C E R N S
M I C R O S E R V I C E S A R E …

16. M I C R O S E R V I C E S
A R E N O T L AY E R E D
A R C H I T E C T U R E S
B AT C H D ATA P R O C E S S
D ATA A C C E S S L AY E R
P R E S E N TAT I O N T I E R
M E S S A G E Q U E U E

17. B U S I N E S S
C O N C E R N S
B AT C H D ATA P R O C E S S
D ATA A C C E S S L AY E R
P R E S E N TAT I O N T I E R
TaxPricing

18. I N D E P E N D E N T LY
D E P L O YA B L E
M I C R O S E R V I C E S A R E …

19. B U S I N E S S
C O N C E R N S
B AT C H D ATA P R O C E S S
D ATA A C C E S S L AY E R
P R E S E N TAT I O N T I E R
TaxPricing

20. B U S I N E S S
C O N C E R N S
B AT C H D ATA P R O C E S S
D ATA A C C E S S L AY E R
P R E S E N TAT I O N T I E R
Tax v2Pricing

21. T H E
F R A C T U R E D
M O N O L I T H
B E WA R E …

22. C O N TA I N I N G
A L L Y O U R
L O G I C
A M I C R O S E R V I C E H A S …

23. - M A R T I N F O W L E R
“Smart Endpoints, Dumb Pipes”

24. S M A R T P I P E S

25. H O W D O E S T H I S M AT C H
O U R “ I D E A L A P I ” ?

26. D E C O U P L E D
• Implementation can’t be
leaked due to separate
deployments
• Corollary:
Microservice
architectures can be
polyglot
• All the intelligence is in the
system

27. R E U S A B L E
• Services contain “just
enough” information to be
useful in multiple places
• No danger of long strands
of dependencies

28. S C A L A B L E
• Separate systems can each
be deployed
independently of the other
• Smaller systems have less
requirements in terms of
hardware

29. A N E X A M P L E !

31. • Youth-targeted BBC news
site
• Targeting a range of
devices
• Needing to deal with a
large amount of traffic and
spikes

32. S 3
N E W S B E AT ‘ M I C R O S E R V I C E ’
A R C H I T E C T U R E
B B C
C O N T E N T
S E R V I C E S
F R O N T- E N D
F E T C H &
D I S P L AY
L B
SQS
R E N D E R E R S

33. • Robust
• Encapsulated and
Decoupled
• Extensible

34. W H AT ’ S T H E
D O W N S I D E ?
Q U E S T I O N 3 :

35. M I C R O S E R V I C E S A R E N O T
R I G H T F O R E V E RY O N E
WA R N I N G

36. M O R E
R E Q U E S T S
• Increased latency
• Less suitable for mobile
platforms

37. M O R E
C O M P L E X
• Upfront costs in terms of
set-up and maintenance
• Architecture can be more
complex
• Issues are less
discoverable

38. C O S T
• Added complexity
• Repeated licensing costs
for non-open source
platforms

39. B U T…
• Similar issues exist with
any “encapsulation”
methodology:
• OOP
• Web Components

40. H O W C A N I S TA R T T O
I M P L E M E N T T H E M ?
Q U E S T I O N 4 :

41. T H E R E ’ S T O O M U C H T O
TA L K A B O U T I N O N E TA L K .
WA R N I N G

42. C R E AT I O N
S T E P # 1 :

43. C R E AT I O N
• Do it yourself!
• Lightweight languages
such as NodeJS make
API creation easy
• App creation
frameworks like Spring
Boot (Java) and
Yeomen (JS)

44. C R E AT I O N
• DropWizard

45. C R E AT I O N
• Automated API
Generation
• e.g. Swagger,
Strongloop

46. P O P U L AT I O N
S T E P # 2 :

47. J U S T C O D E I T !
P O P U L A T I O N

48. T H I N G S C A N
B E H A R D E R
B U T…

49. S O W H E R E D O
W E F O C U S ?
• Security
• Greater documentation
• Integration testing

50. A L S O T E A M
C H A N G E
• “organizations which
design systems ... are
constrained to produce
designs which are copies
of the communication
structures of these
organizations"
• - Conway’s Law

51. M E S S A G E
PA S S I N G
S T E P # 3 :

52. M E S S A G E
PA S S I N G
• “Smart Endpoints, Dumb
Pipes” - Martin Fowler
• All the standard ways of
consuming APIs - e.g.
HTTP, Message Queues

53. D E P L O Y M E N T
S T E P # 4 :

54. D E P L O Y M E N T
• Five years ago:
• Work hard

55. D E P L O Y M E N T
• Automate it as much as
possible
• Old favourites: Continuous
Integration/Source Control
• Puppet/Chef for config
management
• Docker can automate
microservice containers

56. T O C O N C L U D E
S O …

57. T O
C O N C L U D E …
• APIs work best when they
allow separate parts of
your system to behave
independently
• Using microservices can
enforce this pattern
• The Ecosystem is big - and
growing!

58. T H A N K S
@swamwithturtles
swamwithturtles.com