Matinée - Dix conseils pour supporter la croissance de votre Startup de 0 à 10 millions d'utilisateurs Cette session vous a plû ? Si c'est le cas, n'hésitez pas à vous inscrire à notre Summit !

1. 10 Pro Tips for scaling your
startup from 0-10M users
Stephan Hadinger
Sr Mgr, Solutions Architecture
@aws_actus

2. Amazon.com, 1995

9. 2 PIZZA TEAMS

10. Consumer Business
> 270million active
customers accounts
13 countries:
Australia, Brazil, Canada, China,
France, Germany, India, Italy,
Japan, Mexico, Netherlands, Spain,
United Kingdom
Seller Business IT Infrastructure
Business
> 2 million sellers on
Amazon websites
Use Amazon technology for
your own retail website
Leverage Amazon’s massive
fulfillment center network
> 1 million active customers
in over 190 countries
Cloud computing
infrastructure for hosting
web-scale solutions

11. 9 years young
Amazon S3 launched March 14th 2006

13. “AMAZON WEB SERVICES IS PROBABLY THE MOST
IMPORTANT THING THAT HAS HAPPENED TO MOBILE AND
WEB APP DEVELOPERS THAT THE PRESS JUST MISSES.
- Steve Blank, The Four Steps to the Epiphany - The Startup Owner’s Manual
…JEFF BEZOS HAS ACCIDENTALLY OR MAYBE ON
PURPOSE POWERED A WHOLE GENERATION OF
APPLICATIONS.”

14. So when do you get
started?

15. AWS powering a whole generation of startups

16. on
DAY 1

17. Product Development
MVP
Time
Scale

18. Iterative Product Development
MVP
Time
Scale

19. Iterative Product Development
MVP
Time
Scale
Started: burbn, location-based mobile app.
Photo sharing is just one feature
Now: re-written as
photo app. Sold to FB
for 1bn

20. AWS powering a whole generation of startups

21. AWS powering a whole generation of startups

23. http://www.slideshare.net/gueste94e4c/dropbox-startup-lessons-learned-3836587
Pro Tip #1
Learn early, learn often

24. http://www.slideshare.net/gueste94e4c/dropbox-startup-lessons-learned-3836587

25. Pro Tip #2
Not Launching = Painful
Not Learning = Fatal
http://www.slideshare.net/gueste94e4c/dropbox-startup-lessons-learned-3836587

26. “Code something in a day and release it”
AWS Elastic Beanstalk

27. “Code something in a day and release it”
AWS Elastic Beanstalk

28. Speed of iteration
beats quality of iteration

29. 11.6s
Mean time between
deployments
(weekday)
1,079
Max number of
deployments in a
single hour
10,000
Mean number of
hosts
simultaneously
receiving a
deployment
30,000
Max number of
hosts
simultaneously
receiving a
deployment
DEPLOYMENTS AT
AMAZON.COM

30. Pro Tip #3
“Keep the main thing the main thing”
http://www.slideshare.net/gueste94e4c/dropbox-startup-lessons-learned-3836587

31. http://www.slideshare.net/gueste94e4c/dropbox-startup-lessons-learned-3836587

32. “Fortunately, we spent almost all our effort
on making an elegant, simple product that
‘just works’ and making users happy”
Drew Houston, Founder, Dropbox

33. 1. Learn early, learn often
2. Not Launching = Painful, Not Learning = Fatal
3. Keep the main thing the main thing

35. http://aws.amazon.com/solutions/case-studies/airbnb/
Pro Tip #4
Anytime you can use an AWS service to solve a problem, do it
so that you don’t have to solve that problem yourself

36. Total # AirBnB Guests
9MUp by over 5 million since the
beginning of the 2013

37. AWS = building blocks

38. API
Regions Availability Zones Edge Locations
Storage
S3 EBS Glacier Storage
Gateway
Foundation
Services
Networking
VPC Direct
Connect
ELB Route53
Databases
RDS ElastiCacheDynamo RedShift
Content Delivery
CloudFront
Analytics
DataPipelineEMR Kinesis
EC2
Compute
WorkSpaces
AWS Global Infrastructure
Deployment&
Management
IAM Federation
Identity & AccessMonitoring
CloudWatch
Deployment & Management
BeanStalk Cloud
Formation
OpsWork CloudTrail
Libraries, SDK’s
Web Console
Interaction
Human Interaction
Support
Command Line
AWS Global Infrastructure
Application
Services
Application Services
SES SNS SQS Elastic
Transcoder
CloudSearch SWF AppStream
*Non comprehensive list

39. API
Regions Availability Zones Edge Locations
Storage
S3 EBS Glacier Storage
Gateway
Foundation
Services
Networking
VPC Direct
Connect
ELB Route53
Databases
RDS ElastiCacheDynamo RedShift
Content Delivery
CloudFront
Analytics
DataPipelineEMR Kinesis
EC2
Compute
WorkSpaces
AWS Global Infrastructure
Deployment&
Management
IAM Federation
Identity & AccessMonitoring
CloudWatch
Deployment & Management
BeanStalk Cloud
Formation
OpsWork CloudTrail
Libraries, SDK’s
Web Console
Interaction
Human Interaction
Support
Command Line
AWS Global Infrastructure
Application
Services
Application Services
SES SNS SQS Elastic
Transcoder
CloudSearch SWF AppStream

40. Low-cost, fast development on AWS
Time
Scale
Scenario
Small team with initial idea for Mobile app
3 months to get to launch
Unknown customer/problem/solution
No cash….

41. Dev / Test Environment
Time
Scale
Average Spend
$0
p/m
*Included in AWS Free Tier

42. Amazon EC2
Elastic Load
Balancing
Elastic
Virtual Servers
in the cloud
Dynamic traffic
distribution
Amazon
Route 53
Availability
Zone B
Availability
Zone A
Domain Name System
(DNS) web service
Amazon RDS
Managed relational
database service
DBA

43. 1. Learn early, learn often
2. Not Launching = Painful, Not Learning = Fatal
3. Keep the main thing the main thing
4. Use AWS services to solve problems that you don’t have to solve yourself

45. http://aws.amazon.com/solutions/case-studies/flipboard
Pro Tip #5
Focus less on MySQL administration,
more on scaling out the rest of your services

46. Alpha Release
Time
Scale
Average Spend
$15
p/m

47. Beta Release / MVP
Time
Scale
Average Spend
$235
p/m

48. Getting to MVP for $250
Time
Scale
$235$15$0
Total Spend to MVP
$250 • 3 months dev/test/release
• Serving Beta customers
• Ready for full production
and scale

49. 1. Dev & Test
2. Alpha Release
3. Beta Release

50. http://aws.amazon.com/solutions/case-studies/flipboard
Pro Tip #6
Focus on distributed services and fault tolerant
systems from Day 1

52. 1. Learn early, learn often
2. Not Launching = Painful, Not Learning = Fatal
3. Keep the main thing the main thing
4. Use AWS services to solve problems that you don’t have to solve yourself
5. Focus less on MySQL administration, more on the rest of your services
6. Focus on distributed services and fault tolerant systems from Day 1

54. One of the fastest growing sites in history. Cites AWS
for making it possible to handle growth and scale
http://highscalability.com/blog/2012/5/21/pinterest-architecture-update-18-million-visitors-10x-growth.html

55. February 2013
48.7 million users globally
Raised $200M (Total = $338M)
$2.5B valuation
How do we keep costs down as we scale up?

56. Pro Tip #7
Use Auto-scaling
http://highscalability.com/blog/2012/12/12/pinterest-cut-costs-from-54-to-20-per-hour-by-automatically.html

57. EC2 Instance Hours S3 Storage Volume
UP 293% UP 1700%
• 80 million objects stored in S3
• 410 terabytes of user data
• 70 master databases
• 150 EC2 instances in the web tier
• 90 instances for in-memory caching
• 35 instances used for internal purposes
• Elastic Load Balancing
• Elastic MapReduce (Hadoop)

66. Auto-scaling at Pinterest
Switch off 40% of resources
between peak & off-peak
automatically

67. Pro Tip #8
Use “Reserved Instances”
http://highscalability.com/blog/2012/12/12/pinterest-cut-costs-from-54-to-20-per-hour-by-automatically.html

68. AWS offers multiple purchasing models
On-Demand
Pay for compute
capacity by the hour
with no long-term
commitments
For spiky workloads,
or to define needs
Reserved
Make a low, one-time
payment and receive a
significant discount on
the hourly charge
For committed
utilization
Spot
Bid for unused capacity,
charged at a Spot Price
which fluctuates based
on supply and demand
For time-insensitive or
transient workloads

69. • Auto-scaling – automated shut down of 40% of instances off-peak
• Reserved Instances – to save on EC2 for base workload
Savings: 71%

70. 1. Learn early, learn often
2. Not Launching = Painful, Not Learning = Fatal
3. Keep the main thing the main thing
4. Use AWS services to solve problems that you don’t have to solve yourself
5. Focus less on MySQL administration, more on the rest of your services
6. Focus on distributed services and fault tolerant systems from Day 1
7. Use Auto-scaling
8. Use “Reserved Instances”

71. http://aws.amazon.com/solutions/case-studies/supercell/

72. Pro Tip #9
Look into all possible ways to improve
product and user experience
Hint: this involves lots of analytics behind the scene

73. “THANKS TO AMAZON WEB SERVICES, WE CAN DELIGHT OUR PLAYERS WORLDWIDE.”
Sami Yliharju | Services Lead

74. Let’s add
Big Data
for analytics of web, mobile, gaming, and log data
Amazon EMR
(Elastic Map Reduce)
Amazon Kinesis
Hosted Hadoop
framework
Real-time processing of
large, distributed data
streams
Amazon Redshift
Petabyte-scale
data warehouse service

75. http://aws.amazon.com/solutions/case-studies/flipboard
Pro Tip #10
Control your costs as your user base grows

76. Minimizing Costs while scaling to 100M users

77. 1. Learn early, learn often
2. Not Launching = Painful, Not Learning = Fatal
3. Keep the main thing the main thing
4. Use AWS services to solve problems that you don’t have to solve yourself
5. Focus less on MySQL administration, more on the rest of your services
6. Focus on distributed services and fault tolerant systems from Day 1
7. Use Auto-scaling
8. Use “Reserved Instances”
9. Look into all possible ways to improve product and user experience
10. Control your costs as your user base grows

78. Pro Tip #11

79. Pro Tip #11
Have fun, make history

80. Moving Viadeo to AWS
Julien Simon, Chief Technology Officer
jsimon@viadeoteam.com
@julsimon

81. 25
China
9
France
Russia
 JV 50/50 with local partner,
 1m members
 Position in high quality professional
profiles
French-speaking countries
in Africa
 Ideally positioned organically with strong
momentum
 #1 with 3m members
France
 Founded in 2005
 Market leader, with 9m
members
 Very strong ties with local
recruiters
 Strong corporate offering,
with 21% of top 1,000
companies recruiting on
Viadeo
 Acquired in 2008
 #1 with 25m members
 Trusted source of professional data in
China
 Bringing trust into the professional world
 Strong local management team
China
3
Africa
Viadeo

82. Viadeo in 2015 ?

83. Our current platform
• Infrastructure : 250 Linux servers hosted in San Francisco, CA
• Backends : MySQL, ElasticSearch, Hbase, Spark, Hadoop
• Service-oriented platform : « Kasper », Java
– CQRS : Command-Query Responsibility Segregation
– ES : Event Sourcing
– DDD : Domain Driven Design
• Web application : « Limbo », Node.js
• Mobile applications : iOS and Android

84. AWS @ Viadeo : the story so far
• 2011 : file storage (Amazon S3)
• 2012 : Viadeo office flooded! Internal servers destroyed
 All services rebuilt in AWS in 48h (VPC, EC2)
• 2013 : data processing (Hadoop, Amazon EMR)
• 2014 : more data !
– New analytics infrastructure :
Snowplow  S3  Redshift
(≈ 20 million events / day for starters)
– Content personalization : EMR, Spark

85. AWS @ Viadeo : the story so far
• Infrastructure live in 3 regions (us-east-1, us-west-2, eu-west-1)
• 4 VPC
• ≈ 100 EC2 instances (half production, half development)
• ≈ 15 TB in Amazon S3
• 2 Amazon Redshift clusters (5 instances each)
• AWS services currently in production at Viadeo :
– Core infra : VPC, Route 53, IAM, S3, CloudFront,
– Instances : Elastic Load Balancing, Elastic Beanstalk,
CloudFormation, EC2
– Data storage & processing : RDS, EMR, Redshift

86. Enter physical infrastructure challenges…
• Improving our agility
– How can we deploy infrastructure as often as we deploy code (i.e. every day)?
– How can we experiment quickly and at (almost) zero cost?
• Optimizing our spend
– How can we avoid CAPEX peaks caused by hardware refresh?
– How do we best adapt spend to traffic and business conditions?
• Implementing a stronger Disaster Recovery plan
– …without building a 2nd DC which we don’t need anyway
• Scaling storage & CPU for data processing
– Do we really need big servers, lots of rack space and lots of power?
– How do we efficiently handle unpredictable workloads?

87. Why we decided to move everything to AWS
• We want to focus on our real job : building a great service
– No more issues with hardware vendors, no more licensing hell. Ever!
– Experimenting, (re)deploying & scaling are just a few clicks away
– Great ecosystem of SaaS partners running in AWS
• We’ve been using it for a while now and it just works
– AWS CloudFormation is infrastructure as code
– « Time to server » is in minutes, not days
• We’re ready and eager for it : Agile and DevOps are strong Viadeo values
• We want to measure ROI and optimize cost

88. Our AWS manifesto
• Key objectives : automation, scalability, safety
• Continuous integration & delivery on all layers
(infrastructure, instances, applications)
• Infrastructure : deep integration with AWS services for maximum leverage
• Applications : use our current stack for now, unless the benefit/cost ratio of
adopting an equivalent AWS service is too good to pass (Redshift)
• Move everything, but no big bang : the move will be gradual
 plan & build with parallel run in mind (DC + AWS)

89. Infrastructure as code: done!
Github
CircleCI
Jenkins
Amazon S3
AWS AMIs
packer.io
AWS
CloudFormation
Puppet codebase, CloudFormation templates

90. New challenges ;)
• Some performance & cost trade-off in the short term
• Technical debt (aka « guess what’s under the rug? »)
– Layer 7 load balancing rules (*evil*), legacy filers, etc.
– Servers running really old versions of whatever: keep or reinstall?
– Any cruft ignored over the years. Time to clean up!
• Parallel run requires good connectivity AWS DirectConnect
• CloudFront performance not great across regions  need a 2nd CDN
• Negotiating early termination of the legacy hosting contract !

91. Conclusion
• 5 years ago : «Cloud computing? Why?»
• Now : «NO cloud computing? Why?»
• Some good reasons, but mostly really bad
ones
• Cloud computing a fashion? I don’t think so.
• Cloud = infrastructure in digital form
(like photos, music, movies, money, etc.)
• Using AWS helps Viadeo change every day,
by speeding up innovation and delivery
• See you there!

92. Inscrivez-vous gratuitement à l’adresse :
aws.amazon.com/summits/paris