Your SlideShare is downloading. ×
0
DevOp<br />1 Admin vs. 1.000 Serversphpunconf Hamburg 2011<br />Thomas Metschke<br />@tmetschke<br />2011/09/10<br />Perit...
Peritor GmbH<br />Consulting, development and workshops since 2005<br />2<br />Databases<br />Selection<br />Integration<b...
3<br />
EC2 & EucalyptusCluster Management<br />Automatic Configuration<br />Self Healing Clusters<br />Auto Scaling Clusters<br /...
Project life cycle<br />Maturity &Decline<br />Production& Growth<br />Concept & Development<br />Niche /<br />EOL<br />5<...
Cloud Computing<br />6<br />
7<br />
Amazon Elastic Compute Cloud<br />Virtual servers via API call<br />Pay per hour<br />Different instance types in differen...
Regions and Availability Zones<br />Regions<br />Multiple Availability Zones<br />Slightly different costs<br />Availabili...
Server types<br />10<br />
Amazon EC2 additional services<br />11<br />Elastic Load Balancing<br />Virtual Private Cloud <br />Elastic IP<br />CloudW...
Amazon EC2<br />12<br />
13<br />Storage &IP Addresses<br />
Single AZ Deployment<br />14<br />
Multi AZ Deployment<br />15<br />
Amazon EC2 is<br />16<br />Infrastructure as a Service<br />
What you can build fast<br />17<br />
But what if …<br />18<br />
Heal and restore<br />19<br />
Scaling up and down<br />Manual<br />Time<br />API<br />Load<br />20<br />Different triggers make it possible to scale any...
Deployment<br />21<br />Cluster<br />Application<br />Handle deployments of applications out of source control systems or ...
Automationis needed<br /><ul><li>Configuration
Healing
Scaling
Deployment</li></ul>flickr.com/photos/torek/3788181603<br />22<br />
23<br />Chef– short intro<br />
Chef<br />Open Source Provisioning Tool<br />Configures your servers<br />Cookbooks & recipes<br />Ruby DSL<br />24<br />
Chef Solo/Client<br />25<br />Chef<br />Solo<br />JSON<br />push<br />configuration<br />(re)configure<br />Cookbooks<br />
Cookbooks<br />26<br />Cookbooks<br />apache2<br />memcached<br />mysql<br />zend<br />
Memcached Cookbook<br />27<br />Cookbooks<br />memcached<br />attributes<br />recipes<br />templates<br />
Memcached Cookbook<br />28<br />memcached<br />attributes<br />default.rb<br />recipes<br />service.rb<br />install.rb<br ...
Attributes<br />29<br />cookbooks/memcached/attributes/default.rb<br />Set defaultvaluesfor variables to use in recipes & ...
Recipes<br />30<br />cookbooks/memcached/recipes/service.rb<br />Actualinstructions to execute<br />
Recipes<br />31<br />cookbooks/memcached/recipes/install.rb<br />
Templates<br />cookbooks/memcached/templates/default/memcached.conf.erb<br />32<br />
/etc/chef/solo.json    (configuration JSON)<br />Chef Solo Run<br />33<br />Run Chef solo<br />
Let’s talk about cute little Monsters<br />34<br />
Monster World<br />1.6 million DAU<br />8.2 million MAU<br />Peak 130 servers<br />~ 240.000 requests per minute on averag...
Basic architecture<br />36<br />App Server<br />App Server<br />EBS Volume<br />Load Balancer<br />App Server<br />App Ser...
Flexibility and Scalability<br />1:5 facebook<br />1:20 local social networks<br />
Flexibility and Scalability<br />Cost savings of ~ 40%!<br />
Upcoming SlideShare
Loading in...5
×

1 DevOp vs 1.000 servers - Amazon EC2 and Chef automation intro

7,553

Published on

Short intro into Chef and infrastructure automation on the phpunconf in Hamburg, Germany

Published in: Technology, Self Improvement
0 Comments
5 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
7,553
On Slideshare
0
From Embeds
0
Number of Embeds
7
Actions
Shares
0
Downloads
44
Comments
0
Likes
5
Embeds 0
No embeds

No notes for slide

Transcript of "1 DevOp vs 1.000 servers - Amazon EC2 and Chef automation intro"

  1. 1. DevOp<br />1 Admin vs. 1.000 Serversphpunconf Hamburg 2011<br />Thomas Metschke<br />@tmetschke<br />2011/09/10<br />Peritor GmbH<br />
  2. 2. Peritor GmbH<br />Consulting, development and workshops since 2005<br />2<br />Databases<br />Selection<br />Integration<br />Migration<br />Automation<br />Deployment<br />Workflows<br />Clusters<br />Ruby & RoR<br />Code Review<br />Security Review<br />Workshops<br />Development<br />Performance<br />Analysis<br />Optimization<br />
  3. 3. 3<br />
  4. 4. EC2 & EucalyptusCluster Management<br />Automatic Configuration<br />Self Healing Clusters<br />Auto Scaling Clusters<br />One Click Deployment<br />One Click Cluster Cloning<br />Monitoring & Alerting<br />Logs & Accounting<br />Firewalls & Backups<br />Access & Rights Management<br />4<br />
  5. 5. Project life cycle<br />Maturity &Decline<br />Production& Growth<br />Concept & Development<br />Niche /<br />EOL<br />5<br />
  6. 6. Cloud Computing<br />6<br />
  7. 7. 7<br />
  8. 8. Amazon Elastic Compute Cloud<br />Virtual servers via API call<br />Pay per hour<br />Different instance types in different regions <br />8<br />
  9. 9. Regions and Availability Zones<br />Regions<br />Multiple Availability Zones<br />Slightly different costs<br />Availability Zones<br />4 x US-east – Northern Virginia<br />3 x US-west – Northern California<br />3 x EU-west – Ireland<br />2 x AP-southeast – Singapore<br />2 x AP-northeast – Tokyo<br />9<br />
  10. 10. Server types<br />10<br />
  11. 11. Amazon EC2 additional services<br />11<br />Elastic Load Balancing<br />Virtual Private Cloud <br />Elastic IP<br />CloudWatch<br />Spot instances<br />Elastic Block Storage<br />And a lot more …<br />
  12. 12. Amazon EC2<br />12<br />
  13. 13. 13<br />Storage &IP Addresses<br />
  14. 14. Single AZ Deployment<br />14<br />
  15. 15. Multi AZ Deployment<br />15<br />
  16. 16. Amazon EC2 is<br />16<br />Infrastructure as a Service<br />
  17. 17. What you can build fast<br />17<br />
  18. 18. But what if …<br />18<br />
  19. 19. Heal and restore<br />19<br />
  20. 20. Scaling up and down<br />Manual<br />Time<br />API<br />Load<br />20<br />Different triggers make it possible to scale any part of the infrastructure or create new clusters<br />
  21. 21. Deployment<br />21<br />Cluster<br />Application<br />Handle deployments of applications out of source control systems or via http downloads. Allow update code on running servers. User management and GUI.<br />
  22. 22. Automationis needed<br /><ul><li>Configuration
  23. 23. Healing
  24. 24. Scaling
  25. 25. Deployment</li></ul>flickr.com/photos/torek/3788181603<br />22<br />
  26. 26. 23<br />Chef– short intro<br />
  27. 27. Chef<br />Open Source Provisioning Tool<br />Configures your servers<br />Cookbooks & recipes<br />Ruby DSL<br />24<br />
  28. 28. Chef Solo/Client<br />25<br />Chef<br />Solo<br />JSON<br />push<br />configuration<br />(re)configure<br />Cookbooks<br />
  29. 29. Cookbooks<br />26<br />Cookbooks<br />apache2<br />memcached<br />mysql<br />zend<br />
  30. 30. Memcached Cookbook<br />27<br />Cookbooks<br />memcached<br />attributes<br />recipes<br />templates<br />
  31. 31. Memcached Cookbook<br />28<br />memcached<br />attributes<br />default.rb<br />recipes<br />service.rb<br />install.rb<br />templates<br />default<br />memcached.conf.erb<br />
  32. 32. Attributes<br />29<br />cookbooks/memcached/attributes/default.rb<br />Set defaultvaluesfor variables to use in recipes & templates<br />
  33. 33. Recipes<br />30<br />cookbooks/memcached/recipes/service.rb<br />Actualinstructions to execute<br />
  34. 34. Recipes<br />31<br />cookbooks/memcached/recipes/install.rb<br />
  35. 35. Templates<br />cookbooks/memcached/templates/default/memcached.conf.erb<br />32<br />
  36. 36. /etc/chef/solo.json (configuration JSON)<br />Chef Solo Run<br />33<br />Run Chef solo<br />
  37. 37. Let’s talk about cute little Monsters<br />34<br />
  38. 38. Monster World<br />1.6 million DAU<br />8.2 million MAU<br />Peak 130 servers<br />~ 240.000 requests per minute on average<br />35<br />
  39. 39. Basic architecture<br />36<br />App Server<br />App Server<br />EBS Volume<br />Load Balancer<br />App Server<br />App Server<br />DBs<br />App Server<br />App Server<br />EBS Volume<br />App Server<br />Slave DBs<br />App Server<br />All configuration, management and deployment through Scalarium.<br />No other tools needed.<br />S3<br />
  40. 40. Flexibility and Scalability<br />1:5 facebook<br />1:20 local social networks<br />
  41. 41. Flexibility and Scalability<br />Cost savings of ~ 40%!<br />
  42. 42. DEMO?<br />39<br />
  43. 43. Want to try Scalarium<br />SignUp on<br />http://scalarium.com<br />and use it for free as long as you are in test & dev mode<br />Chef examples and guides<br />https://github.com/scalarium/example-cookbooks<br />http://wiki.opscode.com/display/chef/Resources<br />40<br />
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×