Readying your IT Infrastructure for Cloud
- 1. Cloud Computing – Readying Your IT Infrastructure for Cloud Michael Shallcross Distinguished Engineer IBM Global Technology Services Asia Pacific
- 2. Agenda Cloud computing – understanding what’s really different Pragmatic steps towards the cloud
- 3. Cloud Computing is a new consumption and delivery model for IT services Business Perspective IT Perspective A user experience and business model Standardised, self service offerings Rapidly provisioned services Flexibly priced Accessed via the network / internet An infrastructure management and services delivery method Virtualised resources Managed as a single large resource Delivered services with elastic scaling Advanced automation
- 4. Cost savings and faster time to value are the leading reasons why companies consider cloud computing Automated Faster cycle times Lower support costs Optimised utilisation Improved compliance Optimised security End user experience Standardised Easier access Flexible pricing Reuse and share Easier to integrate Virtualised Higher utilisation Economy of scale benefits Lower capital expense Lower operating expense Higher quality services Doing more with less Breakthrough agility and reduced risk
- 5. Cloud computing transforms service delivery Lower cost delivery channel Products standardised for electronic delivery Processes re-engineered and policies defined to allow “zero touch” automation Programs to migrate customers Use of the channel expanded over time
- 6. The future will not be “one size fits all” – multiple deployment models will continue to be required Private Cloud Public Cloud Services Enterprise Data Centre Users Enterprise owned Mission critical High compliancy Internal network Shared resources Elastic scaling Pay as you go Public internet Enterprise Data Centre Managed Private Cloud Hosted Private Cloud Shared Cloud Services 3 rd party operated Enterprise 3 rd party hosted & operated Enterprises 3 rd party owned and operated Standardisation Centralisation Security Internal network Mix of shared and dedicated resources Shared facility and staff VPN access Source: IBM Market Insights, Corporate Strategy Private Cloud Community Cloud Public Cloud
- 7. Cloud computing enables many elements of IT to be delivered and consumed as a services EXAMPLES INFRASTRUCTURE AS-A-SERVICE PLATFORM AS-A-SERVICE SOFTWARE AS-A-SERVICE Servers Networking Storage Middleware Collaboration Financials CRM / ERP / HR Industry applications Data centre Shared virtualised, dynamic provisioning Database Web 2.0 application runtime Java runtime Development tools BUSINESS PROCESS AS-A-SERVICE Employee benefits management Industry-specific processes Procurement Business travel 4 3 2 1
- 8. There is an emerging set of common workloads ready for cloud computing – in both public and private delivery models Analytics Data mining, text mining or other analytics Data warehouses or data marts Transactional databases Business services Customer relationship management (CRM) or sales force automation E-mail Enterprise resource planning (ERP) applications Industry-specific applications Collaboration Audio/video/Web conferencing Unified communications VoIP infrastructure Source: IBM Market Insights, Cloud Computing Research , July 2009. Desktop and devices Desktop Service/help desk Development and test Development environment Test environment Infrastructure Application servers Application streaming Business continuity/ disaster recovery Data archiving Data backup Data center network capacity Security Servers Storage Training infrastructure Wide area network (WAN) capacity
- 9. However many medium-large organisations currently still significantly prefer private clouds over public or hybrid clouds Overall, how appealing are the public, private and hybrid delivery models for your company? 64% 30% Public +113% 64% 38% Hybrid Private +68% Private "Very appealing" or "appealing" "Very appealing" or "appealing" Source: IBM Market Insights, Cloud Computing Research , July 2009. n=1,090 However, adoption of Public Clouds is expected to grow by 26% CAGR between now and 2013* *IDC eXchange, IDC’s New IT Cloud Services Forecast: 2009-2013, p=543, Oct 5, 2009
- 10. Concerns about data security and privacy are the primary barriers to public cloud adoption Percent rating the factor as a significant barrier (4 or 5) Respondents could select multiple items 69% 54% 53% 52% 47% Security/privacy of company data Service quality/performance Doubts about true cost savings Insufficient responsiveness over network Difficulty integrating with in-house IT What, if anything, do you perceive as actual or potential barriers to acquiring public cloud services? Source: IBM Market Insights, Cloud Computing Research , July 2009. n=1,090
- 11. Several major factors are driving cloud computing economics … and it’s potential for transforming IT now Takes repeatable tasks and automates services, helping to reduce IT operation costs and deliver faster Allows clients to “serve themselves”— requiring less support and offering easier access to services Reduces complexity, which means that more automation is possible, helping to reduce IT labour costs Self service Automation of management Standardisation of workloads Provides benefits of scale — if virtualised environments are highly utilised Drives reduced capital requirements Virtualisation of hardware Utilisation of infrastructure Labour leverage Infrastructure leverage
- 12. In this new environment, the data centre must provide a reliable and secure foundation for a new range of cloud services
- 13. There are a common set of steps many organisations go through to condition their infrastructure and make it “cloud ready” Simplified Shared Dynamic Virtualise Automate Reduce infrastructure complexity Reduce staffing requirements Improve business resilience (manage fewer things better) Improve operational costs/reduce TCO Remove physical resource boundaries Increased hardware utilisation Allocate less than physical boundary Reduce hardware costs Simplify deployments Standardised services Dramatically reduce deployment cycles Granular service metering and billing Massively scalable Autonomic Flexible delivery enables new processes and services Self-service Elastic scaling Automatic service metering and billing Industrialised service delivery Economies of scale Self-Service Standardise
- 14. Laying the foundation for a private cloud… Virtualisation Virtualised Applications & Middleware Virtual Clients Virtual Networks Virtual Storage Virtual Servers Physical Layer IBM & Other Storage Networking System x, BladeCenter IBM System z Power Systems Non-IBM Servers Workloads Software Development Test and Pre-Production Self Service Automation Service Management Service Catalog & Automation Request UI Operations UI Provisioning SLA Mgmt Monitoring Security Metering Virtualization mgmt Image mgmt Capacity mgmt Customer Consumer Cloud Service Standards and Governance
- 15. Cloud computing is all about services – not just servers Servers have always been with us Virtualisation is a critical and necessary enabler for cloud computing But… The emphasis on services is the aspect of cloud computing that is most different Service lifecycle Service portal Service catalog Service design Service fulfilment Service monitoring Service pricing Take the customer perspective Focus on the user experience and the service being consumed Gain much deeper insight into the value proposition for cloud computing (and the requirements)
- 16. Where you start determines how quickly you deliver results “ From now on every new project will run on the cloud” “ Let’s target some key services that will enable quick wins” “ Let’s clean up this mess and migrate what we have to the cloud” Pre-integrated cloud platform “ Public” cloud services Pre-integrated cloud platform Workload-optimised private cloud Selected public cloud services Cloud strategy Consolidation Virtualisation Automation Custom private cloud
- 17. Development and Test environments are frequently an ideal workload with which to pilot cloud technologies Source: “Industry Developments and Models – Global Testing Services: Coming of Age,” IDC, 2008, and IBM Internal Reports Potential Benefits Reduce IT labour cost by 50% for configuration, operations, management and monitoring of test environments 75% capital utilisation improvement and significant license cost reduction Reduce test provisioning cycle times from weeks to minutes Improve quality – eliminate 30% of defects that come from faulty configurations 30 per cent to 50 per cent of all servers within a typical IT environment are dedicated to test Most test servers run at less than 10 per cent utilisation, if they are running at all Setting up and taking down test environments is extremely labour-intensive, error prone and slow 30 per cent of all defects are caused by incorrectly configured test environments Testing backlog is often the single largest factor in delaying new application deployments Risk Benefit
- 18. Desktop virtualisation – the next opportunity for desktop transformation Lock down security and compliance Be pandemic ready Optimise the use of thin clients Repurpose and give new life to older PCs Reduce cost and complexity associated with managing end user computing Quickly scale and provision / deprovision end users Dramatically reduce energy costs with thin clients Remove IT tasks from the end user freeing them to be more productive Potential Benefits Significantly reduce desk-side support costs Cut help desk calls by up to 40 per cent Simplify operating system, application and security update process Tighten desktop security and resiliency Use up to 45 per cent less power over traditional desktops and laptops Transform distributed IT operations into a centralised, flexible, secure and highly scalable virtualised desktop environment that is cost-effective to run and simple to maintain
- 19. Amongst the early adopters there are already clear patterns that suggest a practical approach to cloud computing Define cloud strategy and roadmap Assess cloud deployment models, service options and workloads Plan cloud strategy and roadmap Choose initial project Start with an isolated private cloud deployment Choose low-risk workload such as test and development Standardise applications and systems Deploy self-service portal Roll out cloud across the enterprise Enable additional workloads on private cloud Add new users Use trusted public cloud services to supplement data centre capabilities Pilot and Deploy Extend and Evolve Plan and Prepare Condition the existing infrastructure for cloud Virtualise and automate existing systems Add service management, service catalog
- 20. Learning points Cloud computing is a new consumption and delivery model for IT-based services with two main motivations: Cost reduction and business optimisation In future there will be three delivery models: private cloud and public cloud co-existing with the traditional enterprise IT delivery model But “one size fits all” doesn’t apply. Workload characteristics will drive cloud adoption strategy. Workload analysis and prioritisation is required. Cloud economics is driven by the level of virtualisation , standardisation , automation and self-service . Understand your current and target maturity in these key areas. Don’t “wait for cloud to mature” Condition the existing infrastructure for cloud Define your cloud strategy and roadmap Start with an isolated private cloud deployment Roll out cloud across the enterprise
- 21. Thank You For more information, please visit: ibm.com/cloud