A Study on Amazon Web Services and Google App Engine


1
A Study on Amazon Web Services and Google App
Engine

• Alan Tran
• EECS 221
• Fall 2008

2
Outline

• What is cloud computing?
• Amazon Web Services
• Google App Engine
• Comparison studies user perspective

3
Cloud Computing

• Increase capacity or add capabilities on the fly
  without investing in new infrastructure or
  licensing new software
• Capabilities provided as a service
• Accessible through Internet
• Utility computing offering storage and virtual
  servers accessible on demand

4
Amazon Web Services

• Amazons infrastructure auto scaling load
  balancing
• Elastic Compute Cloud EC2 scalable virtual
  private server instances
• Simple Storage Service S3
• Simple Queue Service SQS messaging
• SimpleDB - database
• Flexible Payments Service Mechanical Turk
  CloudFront etc.

5
AWS Project

• Use S3 EC2 SQS to create a scalable video
  converter web application
• Use computing power of EC2 virtual server
  instances for video conversion
• Ubuntu image installed/configured FFmpeg
• S3 for media storage
• SQS for messaging between client and EC2
• Ex converting for iPod Youtube

6
AWS Project

• Video converter architecture

7
AWS Project
8
Google App Engine

• Build/host web apps on Googles infrastructure
  auto scaling load balancing
• Datastore version of proprietary DB system
  BigTable
• GQL like SQL but no support for relational DBs
• Webapp framework SDK
• Services Google accounts mail URL fetch image
  manipulation

9
GAE Project

• Video conversion web app not possible
• Alternative image upload and manipulation
• Upload images to Datastore perform auto-resizing
  into thumbnails

10
GAE Project
11
AWS vs. GAE

• AWS
• Very flexible lower-level offering closer to
  hardware more possibilities higher performing
• Runs platform you provide machine images
• Supports all major web languages
• Industry-standard services move off AWS easily
• Require much more work longer time-to-market
• Deployment scripts configuring images etc.
• Various libraries and GUI plug-ins make AWS do
  help

12
AWS vs. GAE

• GAE
• Very tightly controlled Python only no
  installation/config of open source software
• Proprietary environment hard to move away from
• Easy to use free but limited
• fees coming soon
• Further support may be limited ex Ruby on Rails
  tightly coupled with relational DBs

13
Should I use AWS or GAE?

• Use AWS if you
• Want to use third party open source software
• Have existing code
• Want to transfer web app to own machine/servers
  later on
• Port code to another language
• Want control EC2 instances and machine images
• Ex need to stress/load test app just load up
  1000 instances
• Features messaging payment services etc

14
Should I use AWS or GAE?

• Use GAE if you
• Have no pre-existing code
• Want web app to go live as soon as possible
• Are satisfied with Python GAE framework
• Not doing anything fancy installing software
• Are satisfied with quota restrictions until they
  make you pay

15
Pricing Comparison Example

• Upload/convert 100 5MB videos from MP4 to FLV and
  serve all videos once a day for a month
• Misc. hosting plans 30-50 with downgrade in
  service slower conversion time
• AWS 85 with proven dependable service and
  scalability
• Self-hosting purchase servers storage
  solutions time spent
• As site increases with visitors hosting plans
  can go up to 500/month AWS will save more

16
Conclusion

• AWS and GAE extremely scalable reliable
  platforms for web apps
• AWS has definite edge
• Not many sites on GAE now
• Google adding improvements and language support

17
References

• http//aws.amazon.com
• http//appengine.google.com
• http//developer.amazonwebservices.com/connect/ent
  ry.jspa?externalID691
• Programming Amazon Web Services