6. Cloud Confusion “The interesting thing about cloud computing is that we’ve redefined cloud computing to include everything that we already do. I can’t think of anything that isn’t cloud computing with all of these announcements. The computer industry is the only industry that is more fashion-driven than women’s fashion. Maybe I’m an idiot, but I have no idea what anyone is talking about. What is it? It’s complete gibberish. It’s insane. When is this idiocy going to stop?” - Larry Ellison, 9/08 6
10. From Electricity to IT “The bulk of business computing [will shift] out of private data centers and into the cloud.” - Nicholas Carr, The Big Switch 10
12. Cloud Computing According to the Analysts “A pool of highly scalable, abstracted infrastructure, capable of hosting end-customer applications, that is billed by consumption.” - Forrester “A style of computing where massively scalable IT-related capabilities are provided ‘as a service’ across the Internet to multiple external customers.” - Gartner 12
13. Cloud Computing According to NIST “Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.” http://csrc.nist.gov/groups/SNS/cloud-computing/ 13
14. Five Characteristics of Public Clouds Broad Access Lightweight entry/exit On-demand Usage-based pricing Highly scalable/elastic 14
17. Application Clouds (SaaS) Ease of Use Low Complexity Flexibility Minimal Control Typical Consumers End Users Examples Salesforce.com TurboTax Online Microsoft Online Services Rackspace E-mail 17
18. Platform Clouds (PaaS) Ease of Use Medium Complexity Flexibility Medium Control Typical Consumers Developers Examples Rackspace Cloud Sites Google AppEngine Force.com 18
19. Infrastructure Clouds (IaaS) Ease of Use High Complexity Flexibility Maximum Control Typical Consumers Developers System Administrators Examples EC2, S3 Rackspace Cloud Files, Cloud Servers FlexiScale GoGrid 19
21. Cost Savings From Fixed to Variable Pricing Traditional: Buy enough compute to satisfy maximum anticipated demand Cloud: Dynamically buy enough compute to satisfy actual demand Cost savings particularly significant for transitory compute needs 21
22. Cost Savings Example Problem NY Times wanted to convert 11 million articles from 1851-1922 from raw TIFF images (4 TB) to PDF Solution Leveraging 100 EC2 instances and S3, all 11 million articles were converted to PDF in just under 24 hours Compute Cost? $240 http://open.blogs.nytimes.com/2007/11/01/self-service-prorated-super-computing-fun http://aws.typepad.com/aws/2008/02/taking-massive.html 22
23. Time Savings From Days/Weeks to Minutes Traditional: Procure, receive, unpack, rack, cable, configure new server Cloud: On-demand compute via programmatic call or mouse click Essentially eliminate new server deployment latency Significant time to market implications 23
24. Time Savings Example: New Server Deployment Traditional 72,000 minutes or ~7½ weeks * Cloud 2 minutes * Percent Time Savings 99.997% * Actual data from a pharmaceutical enterprise 24
25. Time Savings Example: Batch Processing Assume a job has 100 units of work and 1 server can complete 1 unit/hr at a cost of $1/hr 1 server would take 100 hours to complete the job at a cost of $100 10 servers would take 10 hours to complete the job at a cost of $100 100 servers would take 1 hour to complete the job at a cost of $100 25 Limiting Factor? Capacity
26. Increased -ilities More Agility, Scalability, Flexibility… Traditional: Capacity planning (maybe) to stay ahead of IT resource requirements. Generally unable to meet immediate increases in demand. Cloud: Scale up (and down!) on-demand Virtually unlimited compute and storage available Enables “just enough” compute so you are only using what you need (green) Enables “just in time” compute to auto-scale applications in response to spikes in demand Generally, no commitments Interesting implications for Dev/QA environments, upgrades, and troubleshooting 26
27. Increased Scalability Example YouDecide2008.com In January 2008, information was posted about the presidential debates resulting in an increase from 25,000 visitors to more than 300,000 – in one day. Traditional Hosting = Site Crash! 27
31. The BIGGEST objection… Security! The cloud is for everyone but not for everything. (but remember, cloud doesn’t necessarily mean shared) 31
32. Other Reasons Not To Use the Cloud Reliability? Madden Syndrome: driving vs. flying Performance/lack of isolation concerns Latency Need more control Regulatory issues - HIPPA, SOX, PCI Want a “managed” solution If it ain’t broke, don’t fix it! Some items derived from this IT Management & Cloud blog post 32
35. Closing Thoughts Cloud is still hosting and hosting requires trust “As a Service” should come with service Move one or more non-critical systems to the cloud to get experience Consider hybrid solutions Experiment with backup and archiving to cloud storage It’s still early in cloud time, but... 35
Marc Andreessen definition of “platform” - a system that can be programmed and therefore customized by outside developers -- users -- and in that way, adapted to countless needs and niches that the platform's original developers could not have possibly contemplated, much less had time to accommodate.