Leveraging the Cloud September 22, 2011 Digital Government Institute Cloud-Enabled Government Conference Washington, DC
General Dynamics Information Technology Aerospace Combat Systems $29.3 billion in annual revenue 92,300 employees worldwide Leading market positions in: Mission-critical information systems and technologies Land and expeditionary combat systems, armaments and munitions Shipbuilding and marine systems Business aviation Marine Systems 2
Cloud Computing 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 Characteristics: On-demand Self-service, Broad Network Access, Resource Pooling, Rapid Elasticity, Measured Service Service Models: Infrastructure as a Service, Platform as a Service, Software as a Service Deployment Models: Private, Community, Public, Hybrid How Do We Leverage the Technology? 3
How different does it look from traditional IT CPU CPU CPU CPU Self Serve Portal Machine Machine Machine Appliance Appliance Appliance App Server DB Server FW/LB Web Tier VLAN App Tier VLAN DB Tier VLAN 4
Where do opportunities exist for Cloud Computing that provide great value to the enterprise Disaster Recovery look at insurance model and for those systems and applications that are less critical consider oversubscribe model further driving down cost Technology Refresh pristine image and isolated platform providing shadow for testing new version and patches for compatibility Research and Development (R&D) quick stand-up, sharing and collaborating across enterprises, constant demand changes for computing resources, quick teardown and repurpose Software development model production early on, and quicker standup of environment through agile infrastructure Batch processing not real-time, not end user focused, typically require completion by specified time but not necessarily starting at a fixed time Inter-Agency Development &Collaboration difficult to fund and support needs of all stakeholders through traditional acquisition and O&M Public facing content consider dynamic scalability to support peak times and also stop external users from accessing your network 5
Traditional approach for disaster recovery Scenario: Disaster Recovery Primary Site Primary Data Standby Data Warm/Hot DR Site Replicated Transactions $$$$$$ Warm Site Leased or rented facility with operational infrastructure Supports longer recovery times (vs. hot) May have some duplicate or specialty equipment stored on site Relatively short standup time No initial staff commitment Lower CapEx initially Not immediately available No opportunity to test recovery process Vulnerable to technology obsolescence Extensive cost after recovery initiation Leased or rented facility fully configured and ready to support operations within hours Opportunity for recovery testing exercises Hot Site Pros Cons Pros Cons Short-term residency (expensive) Platform and applications must be fully compatible with primary site Long stand-up time (weeks/months) Continual technology refresh needed Expensive for recovery testing exercises and compliance Extensive CapEx and staff commitment 6
A new approach to COOP and DR can be considered in many environments through virtualization and Infrastructure as a Service Scenario: Disaster Recovery Primary Site App File Server IaaS DR Site Hypervisor Database Database Database Hypervisor App Server App Server Desktop Desktop Desktop Compute Compute Encrypted Replicated Data Desktop Desktop Compute Compute Hypervisor Database Desktop Ease of acquisition Point-n-click self provisioning Fully OpEx funded No CapEx funding Flexible cost models: o Subscription based reservation o Pay for what you use when you use it, for as long as you use it o Flexible SLAs Extremely fast standup time (hours/ days) Fast track to compliance COOP & Disaster Recovery Pros Always available Sync d data, fast cutovers (minutes/hours) Secure Confidentiality, integrity, availability Open-ended recovery assurance testing: On demand testing with multiple scenarios Iterative process refinement per scenario Restore to pre-test state from snapshots Ideal, low cost strategy for both secondary and tertiary sites No technology refresh required No dedicated staff required Cons x86 machine architectures only no specialty equipment or peripherals Windows, Linux or Solaris OS only no specialty operating environments No reserved physical resources SLAs only No custom hardware configurations 7
Comparing Cloud and Traditional DR Scenario: Disaster Recovery Application Layer Middleware & 3 rd Party tools Operating System ization Servers Storage Network Connectivity Traditional Considerations Migrate applications and data Develop and test failover scenarios Configuration management Configure middleware and other tools for synchronization, failover, HA, active-active, etc Leverage images where available Install and configure OS per version description Rack, stack, engineering, configuration, ongoing O&M Capacity Planning Establish point to point connectivity Cloud Considerations SAME SAME SAME Agile infrastructure provides quick provisioning across large portion of technology stack Self service portal provides configuration ability without the need to directly configure hardware infrastructure available without engineering required to design and standup Operating system licensing included with service Identify approach and security Determine bandwidth transfer cost 8