410 likes | 532 Views
Cloud Computing: Getting Beyond the Hype. Linda Y. Cureton CIO, National Aeronautics and Space Administration Tuesday May 4, 2010. Agenda. About NASA Definition of Cloud Computing Cloud Computing and NASA Getting Beyond the Hype. The NASA Mission. The NASA Mission.
E N D
Cloud Computing: Getting Beyond the Hype Linda Y. CuretonCIO, National Aeronautics and Space AdministrationTuesday May 4, 2010
Agenda About NASA Definition of Cloud Computing Cloud Computing and NASA Getting Beyond the Hype
The NASA Mission Aeronautics: pioneers and proves new flight technologies that improve our ability to explore and which have practical applications on Earth. Exploration Systems: creates capabilities for sustainable human and robotic exploration. Science: explores the Earth, solar system and universe beyond; charts the best route of discovery; and reaps the benefits of Earth and space exploration for society. Space Operations: provides critical enabling technologies for much of the rest of NASA through the space shuttle, the International Space Station and flight support. NASA's mission is to pioneer the future in space exploration, scientific discovery, and aeronautics research.
The NASA Mission Glenn Research Center Goddard Space Flight Center NASA Headquarters Ames Research Center Langley Research Center Dryden Flight Research Center Jet Propulsion Laboratory Kennedy Space Center Marshall Space Flight Center Stennis Space Center; NASA Shared Services Center Johnson Space Center 5
The NASA Mission Spinoff Technology: NASA has been issued over 6,300 patents, nearly one in a thousand of all patents ever issued by the U.S. Patent Office (since 1790). Modeling Innovations Advance Wind Energy Industry: In 1981, Glenn Research Center scientist Dr. Larry Viterna developed a model that predicted certain elements of wind turbine performance with far greater accuracy than previous methods. Image-Capture Devices Extend Medicine’s Reach: NASA’s Advanced Diagnostic Ultrasound in Microgravity (ADUM) experiment led to the development of revolutionary medical ultrasound diagnostic techniques for long-distance use, including frame-grabber and data archiving technology that enables ultrasound users with minimal training to send diagnostic-quality ultrasound images and video to medical professionals via the Internet .
NIST Definition Cloud Computing: 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.
Cloud Attributes There are five key Cloud attributes: • Shared / pooled resources • Broad network access • On-demand self-service • Scalable and elastic • Metered by use
Shared / Pooled Resources Resources are drawn from a common pool Common resources build huge economies of scale Common infrastructure runs at high efficiency
On-demand Self-service Completely automated Users abstracted from the implementation Near real-time delivery (seconds or minutes) Services accessed through a self-serve web interface
Scalable and Elastic Resources are dynamically-allocated between users Additional resources are dynamically-released when needed Fully automated
Metered by Use No up-front initial investment is required to get started Users pay only for services used Underlying hardware costs are irrelevant
Benefits of Cloud Computing Reduces Costs No need to purchase hardware, licenses or implementation services Pay-as-you-go Cancel any time
Benefits of Cloud Computing Saves Time Spend less time on IT and more time on business-critical activities Get started immediately Improves workflow through easy sharing and collaboration
Benefits of Cloud Computing Provides Flexibility Access to multiple technologies, business solutions and providers Access services from anywhere in the world Add, subtract or change services as needs change
Benefits of Cloud Computing Greater Resource Efficiency Don’t pay for resources you are not using Just-in-time capacity management Pooled resources saves energy
Benefits of Cloud Computing Higher Quality Service Increased reliability (24/7 availability) Always have the latest software Updates are automatic and immediately available
Benefits of Cloud Computing for Government Automatic policy compliance Faster procurement Faster deployment of technologies Allows Government agencies to focus efforts on strategic projects Prevents duplication efforts Significant cost savings Flexible resources Integrated Reporting
Cloud Models There are three primary Cloud models: • Public Clouds • Private Clouds • Hybrid Clouds
Public Clouds Hosted at a Service Provider Site Supports multiple customers Often utilizes shared infrastructure Supports connectivity over the internet Suited for information that is not sensitive Usually cheaper than a private cloud
Private Clouds Hosted at an Enterprise or Service Provider site Supports only one customer Dedicated infrastructure Connectivity over private network/fiber or the internet May be used for sensitive information More expensive than a public cloud
Hybrid Clouds Hosted at an Enterprise or Service Provider site Supports only one customer Dedicated infrastructure Connectivity over both private network/fiber and the internet May be used to move information between private and public clouds
Cloud Service Models There are three Cloud service models:
Software-as-a-Service • Target Customer: End-User • Provides: • On-demand access to a wide range of Office and Productivitysoftware such as Email, blogs, wikis, forums, etc. • Web Hosting (Google Sites) • Source code control • Key Attributes: • Single or multi-tenant model • Centralized feature and system upgrades • Mobility, flexibility, wide range of options • Can include almost any IT service • Examples: Salesforce.com, Microsoft Federal BPOS, Gmail, Netsuite, ADP, Yahoo email
Platform-as-a-Service • Target Customer: Software Developer • Provides: • Environment in which applications can be developed, tested, staged, deployed, hosted and run • Backup and Fail-over Environments • Standard protocols and common technologies for web app development • Programmable User Interface, Integrated Content Libraries, Real-time analytics & web applications, real-time workflow and approvals, Granular Security & Sharing • Key Attributes: • Multi-tenant deployment architecture • “Infinitely scalable” – one to millions of users • Interacts at the Source Code level • Total abstraction from infrastructure • Examples: force.com, Microsoft Azure, Google App Engine, Yahoo APIs
Infrastructure-as-a-Service • Target Customer: System Administrators • Provides: • Bandwidth, Networks, Operating System, File System, Servers & Storage • Backup power, networks and failover sites • Dedicated team of experts monitoring the system on a 24/7 basis • Key Attributes: • Implementation is hidden, changeable • Often uses Virtualization and Shared Storage (SAN or Cluster) • Similar to utility Computing or Grid Computing • Examples: Amazon EC2, S3 and EBS, Linode, Rackspace Cloud, Terramark Enterprise Cloud, IBM CloudBurst, Nebula IaaS
2010 Federal Budget “The Federal technology environment requires a fundamental re-examination of investments in technology infrastructure… Pilot projects will be implemented to offer an opportunity to utilize more fully and broadly departmental and Agency architectures to identify enterprise-wide common services and solutions, with a new emphasis on cloud-computing.”
Federal Cloud Platform Federal agencies have similar needs Common set of languages, tools, software & services allows us to avoid re-inventing the wheel Standards mitigate vendor lock-in Improves cohesion of citizen experience Easy collaboration and communication between agencies and with private institutions
Federal Cloud Pilots NASA’s Nebula Cloud Computing Platform DISA’s Rapid Access Computing Environment (RACE) National Business Center Cloud Department of Energy’s Magellan
NASA Nebula NEBULA Cloud Computing Platform 2010
NASA’s Data Sharing Challenges NASA has a long history of publishing data Massive data sets Unique formats Difficulty sharing data without granting access to internal networks Open Government Directive
NASA’s IT Challenges • Large-scale infrastructure requirements • Too much money is spent on infrastructure • Fluctuating needs: • Missions Completely Succeed (Rovers) • Missions do not succeed (Orbiting Carbon Observatory) • Politics impact Missions (Triana) • Missions are focused on the Mission • Scientists are focused on the Science
NASA’s Web Challenges Many different web platforms on thousands of sites Widely-fluctuating needs Lock-in, portability, and interoperability issues Confusing to outside users Very large attack surface Very high operating costs Long provisioning times
Nebula: Built for Science • Science-Class Cloud Computing • High CPU-to-Disk Ratio • 10gig E networking • 100 terabyte file system storage • Built for Research • MAE-West Peering (Internet2, NLR, CENIC, 11 Tier-1 ISPs) • Massively Parallel, Loosely Coupled
Nebula: Built for the Web Designed for Search Engines, RSS, and aggregation Friendly URLs It’s RSSable, Tweetable
Nebula Pilot Projects General Services Administration Microsoft World Wide Telescope Google Earth Planetary Content (Mars) Lunar Mapping Module Program (LMMP) Data Processing (Moon)
Nebula Pilot Examples • Data Processing • LMMP uses Nebula to perform stereographic reconstruction using Lunar Reconnaissance Orbiter (LRO) images
Nebula Pilot Examples • Data Sharing • Microsoft World-Wide Telescope • Data for Moon and Mars (100TB+) is processed and hosted on Nebula
Beyond the Hype • The Bull • Technology is so compelling that most marketing strategies claim cloud capabilities • Agree on definition • Know what you need • Don’t let the hype fool you into disinterest • The Silver Bullet • Some applications are not well-suited for cloud • Vendor lock-in • Perceptions on security • The Silver Lining • Look for cloud opportunities • Provision internal clouds as a pilot • Develop cloud-centric solutions
Beyond the Hype “Infrastructure and operations professionals can try to ignore it as it is just in its infancy, but doing so may be a mistake as cloud computing is looking like a classic disruptive technology” - Forrester Disruptive technology should not be ignored Your customers will do this with or without you Follow sound risk-based principles
NASA Nebula Cloud Computing for a Universe of Data http://nebula.nasa.gov @NASAnebula Nebula: NASA’s Cloud Computing Platform NEBULA Cloud Computing Platform 2010