Google Cloud Platform Live - Keynote

from Google Cloud Platform Live 2014
YouTube Video: https://www.youtube.com/watch?v=qokEYBNWA_0

Transcript

1. Keynote, Urs Hölzle
2. Cloud is still too hard Highlights from last year...
3. May 2013 Google Compute Engine in Preview PHP for App Engine in Preview The Last Year in Compute November 2013 1 Million RPS on Compute Engine Load Balancing August 2013 Layer 3 Load Balancing December 2013 Compute Engine GA
4. The Last Year in Storage May 2013 Cloud Datastore August 2013 Encryption at Rest for Google Cloud Storage June 2013 Offline Disk Import October 2013 Native Connections for Cloud SQL December 2013 Persistent Disks February 2014 Cloud SQL GA
5. And a whole lot more... March 2013 Big JOIN in BigQuery February 2014 HIPAA Support July 2013 Dedicated Memcache September 2013 Major BigQuery Updates November 2013 Cloud Endpoints GA Dedicated Memcache GA October 2013 Mobile Backend Starter for iOS
6. Demo Live Migration in Google Compute Engine
7. Source: Google Internal Data 4.75 Million active applications
8. 28 Billion requests per day on App Engine App Engine Wikipedia Popular Site Average Server 86M 500M 2.6B 28B Source: Google Internal Data Source: Google Internal Data
9. 6.3 Trillion Cloud Datastore operations per month Source: Google Internal Data Source: Google Internal Data
10. “Google was the fastest overall. The three Google instances completed the benchmarks in a total of 575 seconds, compared with 719 seconds for Amazon and 834 seconds for Windows Azure. A Google machine had the fastest time in 13 of the 14 tests… Google was also the cheapest overall.”
11. Great Customers
14. Bobby Murphy CTO and Co-Founder, Snapchat
15. Bobby Murphy CTO and Co-Founder, Snapchat
16. Cloud is still too hard Cloud is still too hard
17. Pricing changes Pricing is Complex
18. Pricing changes Pricing is Complex Current clouds force design trade-offs
19. Prices are falling • Public cloud prices have dropped 6-8% annually Source: Google Internal Data 2014 2006 Public Cloud Prices
20. But prices are not falling fast enough • Hardware costs have dropped 20-30% annually Hardware Cost Public Cloud Prices • Public cloud prices have dropped 6-8% annually Source: Google Internal Data 2014 2006
21. Pricing changes On Demand Price Reductions
22. Pricing changes On Demand Price Reductions Sustained-Use Discounts
23. Cloud is still too hard On demand price reductions
24. Google Compute Engine All Regions | All Sizes | All Classes 32% price drop
25. Google Cloud Storage 68% price drop $.026 per GB | $.02 per GB DRA
26. Google BigQuery 85% price drop $5 per TB for on demand usage
27. • Introducing new model with assured and predictable capacity • Easily purchase capacity from 5 to 50GB per second BigQuery
28. Current Industry Pricing Google BigQuery $80,000+ Source: Google Internal Data
29. Current Industry Pricing New Monthly Pricing Google BigQuery $20,000 $80,000+
30. Google BigQuery than the alternatives 75% less expensive Source: Google Internal Data
31. To recap... 32% 68% 85% price drop on Cloud Storage price drop on BigQuery price drop on Compute Engine
32. The original promise of the cloud: low cost, on-demand computing with no need to provision or reserve
33. $0 $500 $1000 $1500 $2000 $2500 $3000 Total cost over 3 years Much more complicated 0% 100% 20% 40% 60% 80% % of 3-year period the instance is running Price On Demand Pricing Reservation Pricing
34. You should get the best price with... No Upfront Payments No Lock-in No Complexity
35. Announcing Sustained-Use Discounts
36. Sustained-use discounts $0.11 $0.10 $0.09 $0.08 $0.07 $0.06 $0.05 $0.04 $0.03 100% 0% 20% 40% 60% 80% Previous On Demand New On Demand Sustained Use Net Price Per Hour
37. 100% 0% 20% 40% 60% 80% Sustained Use Previous On Demand New On Demand $0.11 $0.10 $0.09 $0.08 $0.07 $0.06 $0.05 $0.04 $0.03 Sustained-use discounts Net Price Per Hour
38. Sustained-Use Pricing 53% price drop on Compute Engine instances with 24x7 use
39. To recap... price drop on Cloud Storage price drop on BigQuery price drop on Compute Engine 30-53% 68% 85% Automatic discounts for sustained usage
40. Help realize the original promise of cloud computing Predictable pricing Low on-demand pricing Pricing Principles Simplified pricing Moore’s Law pricing
41. Cloud is still too hard Cloud is still too hard
42. Cloud is still too hard Developers often make trade offs to work around the weaknesses and limitations of today’s public clouds
43. Cloud is still too hard Time to Market or Scalability
44. Cloud is still too hard Time to Market or Scalability Flexibility or Automatic Management
45. Cloud is still too hard Big Data or Real Time Time to Market or Scalability Flexibility or Automatic Management
46. Cloud is still too hard Big Data or Real time Time to Market or Scalability Flexibility or Automatic Management We are changing or to and
47. Greg DeMichillie Director of Product Management
48. Announcing
49. Announcing
50. Announcing Cloud DNS
51. Cloud is still too hard Big Data or Real Time Time to Market or Scalability Flexibility or Automatic Management
52. Developer Productivity Time to Market Scale
53. Developer Productivity Time to Market Scale and
54. Developer Productivity • Use the tools you know and love • • Fast, reliable deployments • • Isolate and fix issues in production Developer Productivity Time to Market Scale and
55. Demo Zero to 60
56. IaaS vs PaaS Flexibility Management
57. and IaaS vs PaaS Flexibility Management
58. Announcing Managed Virtual Machines
59. Managed VMs • Flexibility of Compute Engine with productivity of App Engine • Provides best of both worlds Flexibility Management and
60. Demo Managed VMs
61. $ gcloud app update .
62. Managed VMs: under the hood OS OS
63. Managed VMs: under the hood OS OS Monitoring / Health Checking Task Queues Memcache Cloud Datastore
64. Managed VMs: under the hood Web Server Application Server JVM Web Server Application Server JVM OS OS Monitoring / Health Checking Task Queues Memcache Cloud Datastore
65. Managed VMs: under the hood Web Server Application Server JVM 3rd Party Your Code Web Server Application Server JVM 3rd Party Your Code OS OS Load Balancer Monitoring / Health Checking Task Queues Memcache Cloud Datastore
66. Managed VMs: under the hood Web Server Application Server JVM 3rd Party Your Code Web Server Application Server JVM 3rd Party Your Code OS OS Load Balancer Monitoring / Health Checking Task Queues Memcache Cloud Datastore
67. Demo Managed VMs
68. Big Data Big Data Real Time
69. and Big Data Real Time BigQuery
70. Announcing BigQuery Streaming at 100k Rows Per Second
71. and • Near real-time analysis • High fidelity, low latency • Focus on results, not sharding and transforming Big Data Big Data Real Time
72. Demo BigQuery Streaming
73. BigQuery Streaming • Smart Monitoring • 60 events/hour • Intelligent Network • 400,000 meters • Interactive Analysis • 17,280,000,000 events/month
74. Demo BigQuery Streaming
76. • Build and test in the cloud • AppEngine with Managed VMs • BigQuery Streaming @ 100k RPS • Cloud DNS • Windows Server, SuSE, RHEL support Launches
77. Moore’s Law in the cloud price drop on Cloud Storage price drop on BigQuery price drop on Compute Engine 30-53% 68% 85% Automatic discounts for sustained usage
78. Everything we’re doing is about helping you, the developer, do what you love
79. Cloud Platform Developer Roadshow We’re visiting 27 cities around the world beginning April 7 cloud.google.com/roadshow
80. Thank you