A Tutorial on Hyperscale Infrastructure @ LinuxCon EU 2016

Instructions for the tutorial: https://goo.gl/WYBR6L

Transcript

1. Brandon Philips @brandonphilips | brandon.philips@coreos.com Slides and Demos: https://goo.gl/WYBR6L A Tutorial on Hyperscale Infrastructure Free Stickers @ Podium!
2. 90+ Projects on GitHub, 1,000+ Contributors OPEN SOURCE Secure solutions, support plans, training + more ENTERPRISE CoreOS is Running the World’s Containers Linux
3. Expectations Tutorial
4. Quick Level-Set Follow Along
5. Secure the Internet MISSION
6. Separate Apps from OS STRATEGY
7. Make Servers Consistent STRATEGY
8. Make Servers Easy to Upgrade STRATEGY
9. Simplify Application Management STRATEGY
10. On Any Infrastructure STRATEGY
17. 3 Application packaging Clustering Linux at scale
18. #GIFEE Borg/Omega Linux Chubby
19. #GIFEE Borg/Omega Linux Chubby
20. #GIFEE Borg/Omega Linux Chubby
21. Avoid single points of failure Regularly update software Consistent environment Why build #GIFEE?
22. Why build #GIFEE? Regularly update software
23. Application Packaging 1
24. Containers Run apps consistently
25. Abstract away app from the OS OS App
28. Base software managed by CoreOS kernel systemd OpenSSH
29. Protect apps from each other Isolated network namespace Isolated file system namespace Mixed versions of dependencies eg. python 3.4 & python 2.7
30. $ sudo rkt run coreos.com/etcd:v2.0.0 $ sudo rkt run coreos.com/etcd:v2.0.0 \ --cpu=750m --memory=128M $ sudo rkt run --net=host coreos.com/etcd:v2.0.0 rkt run
31. Search container metadata Identify vulnerabilities Explain update actions Clair container security auditing
32. After scanning millions of containers we found that over 80% still had Heartbleed 80% Clair container security auditing
34. In-Progress Universal Container Format Packaged Downloaded Verified
35. Container Basics DEMO
36. Linux at Scale 2
37. CoreOS Linux Run containers any platform
38. Patches to the OS and kernel are hard Retest after updates No automation SECURITY Dependency breakage Uptime risk APPLICATION
40. Auto-updating browsers fixed security We got HTML5 at the same time
41. Atomic operating system updates
42. Atomic operating system updates
43. CoreOS Linux Run containers any platform
44. Toolbox Tutorial DEMO
45. Clustering 3
46. Patches to the OS and kernel are hard Retest after updates No automation SECURITY Dependency breakage Uptime risk APPLICATION
47. Patches to the OS and kernel are hard No automation SECURITY Uptime risk APPLICATION
48. Operations Paradise Easy scale out Painless app upgrades Tolerant of machine failure
49. App Req/sec: 6,000 App Healthy: True
50. App Req/sec: 6,000 App Healthy: True
51. App Req/sec: 7,000 App Healthy: True
52. App Req/sec: 8,000 App Healthy: True
53. App Req/sec: 7,000 App Healthy: True
54. App Req/sec: 6,000 App Healthy: True
55. App Req/sec: 8,000 App Healthy: True
56. App Req/sec: 7,000 App Healthy: True
57. App Req/sec: 8,000 App Healthy: True
58. App Req/sec: 8,000 App Healthy: True
59. App Req/sec: 8,000 App Healthy: True
60. API for easing scaling, painless upgrades, and failure tolerance
61. Simple cluster operations Secure and Simple API Friendly operational tools db
62. Fault-tolerant database to support complex set of needs Leader election Cluster-wide Semaphores Service discovery Dynamic configuration
63. Hard Computer Science Problem ?
64. Hard Computer Science Problem Chubby
65. Distributed reliable key-value store for the most critical data of a distributed system
66. Etcd Basics DEMO
74. No existing “cloud native” solutions Simple HTTP + JSON APIs Dynamic reconfiguration Why build etcd?
75. Simple key/value “Distributed etc” Keys are versioned Changes can be watch
76. Cluster-wide reboot lock - locksmith Service discovery - vulcand, skydns Cluster orchestration - k8s, cloud foundry
77. Industry Adoption 500+ projects on Github
78. Etcd Clustering DEMO
79. Simple cluster operations Secure and Simple API Friendly operational tools
80. Kubernetes Basics DEMO
81. Scheduler gets work to the servers
82. $ scp app host:/opt $ ssh host systemd-run /opt/app
83. $ scp app host:/opt $ ssh host systemd-run /opt/app
84. $ fab deploy:app
85. $ fab deploy:app
86. $ fab deploy:app
87. $ fab deploy:gpu-app
88. $ fab deploy:gpu-app
89. $ fab deploy:gpu-app
90. App Deployment DEMO
91. ReplicaSets drive the cluster to a desired state
92. pod env=prod app=web pod env=prod app=web pod env=prod app=web rs web-prod select(env=prod,app=web) count=1
93. pod env=prod app=web pod env=prod app=web pod env=prod app=web rs web-prod select(env=prod,app=web) count=1
94. pod env=prod app=web rs web-prod select(env=prod,app=web) count=1
95. App Failure DEMO
96. pod env=prod app=web rs web-prod select(env=prod,app=web) count=5
97. pod env=prod app=web pod env=prod app=web pod env=prod app=web pod env=prod app=web pod env=prod app=web rs web-prod select(env=prod,app=web) count=5
98. App Scaling DEMO
99. Services find scheduled apps (pods) and load balance them
100. pod env=dev app=web pod env=test app=web pod env=prod app=web
101. pod env=dev app=web pod env=test app=web pod env=prod app=web service test.example.com select(env=dev,app=web) service beta.example.com select(env=test,app=web) OR select(env=prod,app=web) service example.com select(env=prod,app=web)
102. pod env=test app=web pod env=prod app=web pod env=prod app=web pod env=dev app=web pod env=test app=web pod env=prod app=web service test.example.com select(env=dev,app=web) service beta.example.com select(env=test,app=web) OR select(env=prod,app=web) service example.com select(env=prod,app=web)
103. pod app=foo,version=1 service foo.cluster.local select(app=foo)
104. pod app=foo,version=1 pod app=foo,version=2 service foo.cluster.local select(app=foo)
105. pod app=foo,version=1 pod app=foo,version=2 service foo.cluster.local select(app=foo)
106. Service Discovery DEMO
107. pod env=prod app=web pod env=prod app=web pod env=prod app=web pod env=prod app=web pod env=prod app=web rs web-prod select(env=prod,app=web) count=5
108. pod env=prod app=web pod env=prod app=web pod env=prod app=web pod env=p rod pod env=prod app=web rs web-prod-4950 select(app=web,hash=4950) count=5 pod env=prod app=web pod env=prod app=web pod env=prod app=web pod env=p rod pod env=prod app=web rs web-prod-1328 select(app=web,hash=1328) count=5 deployment web-prod select(app=web) count=5
109. App Upgrades http://gb.ifup.org DEMO
110. Upstream rktnetes Auth/OIDC Node self-signed TLS
111. Scaling 15x scheduler performance 30k pods on 1k nodes SIG-scale
112. 3 Application packaging Clustering Linux at scale
113. Sounds good, but... Is anyone successful with all this in prod?
114. Publically traded options exchange
115. Containers on CoreOS are powering ISE's high-throughput, low-latency financial exchange Running in production Bare metal & AWS Billions of transactions a day 150 million req/sec
117. Secure the Internet MISSION
118. Separate Apps from OS STRATEGY
119. Make Servers Consistent STRATEGY
120. Make Servers Easy to Upgrade STRATEGY
121. Simplify Application Management STRATEGY
122. On Any Infrastructure STRATEGY
124. Guides & Tools coreos.com/kubernetes
126. Linux
128. Runs on any platform Consistency across clouds Trivial dev, test, & prod Faster time to market
129. Training San Francisco December 8 & 9 New York December 14 & 15 https://coreos.com/training
130. Build, Store and Distribute your Containers quay.io
131. tectonic.com/summit - @TectonicSummit December 12 & 13 2016 - New York City
132. Thank you! Brandon Philips @brandonphilips | brandon@coreos.com | coreos.com We’re hiring in all departments! Email: careers@coreos.com Positions: coreos.com/ careers
133. Bonus: Hacking Setup DEMO