1) The document discusses considerations for building a private cloud, including leveraging the transformational power of cloud computing to enable new business models, deliver IT without boundaries, and improve business agility.
2) It recommends mapping current applications and services to a cloud deployment strategy to prioritize workloads for migration to private, public, or hybrid clouds.
3) The evolution from current infrastructure to a cloud-based delivery model is described, starting with virtualization and advancing to consumption-based metering and automation of service delivery.
Technology will play a critical role in organizations achieving objectives. In fact, in IBM ’s latest CEO study, technology has become the number 1 external factor impacting organizations. Few examples of this…..Cloud speeds service delivery, and so we now see 90% of CIOs view cloud as critical to their plans. We see organizations leveraging mobile to improve reach of services to clients – with 1Bm smart phones and tables shipping in 2012 – a 34% increase in a year. We see over 20B machine centric devices now deployed enabling more responsive business operations and business models. And all of these technologies are generating tremendous data with 2.7 zetabytes of digital content in 2012 – an increase of 50% in one year. How many of you are considering adopting a cloud style delivery approach. Or already have cloud projects?
To many people in lines of business, IT is seen as an issue and not a driver. Lines of business are attracted by the messages of what “cloud” can offer. Agility, flexibility, pay as you go. Doing nothing, or continuing as usual is not an option for IT. Organizations must transform IT from cost centers driving on-going operations to strategic centers of business innovation Today’s business infrastructures are becoming inhibitors to business change. Main point: IT operating costs are rising, IT must break through budget and resource barriers. This image represents a composite of research from IDC, from an IBM commissioned study conducted by Forrester Consulting and from IBM's own Market Insights research. The bottom line is that IT professionals face unnecessary delays and costs as they deploy, maintain and update their capabilities. It's not just an IT headache. It's an expensive IT headache. Research shows on average more than 70 percent of IT budgets are spent on operations and maintenance. Clients have said, "When we do bring new capabilities on board—it takes too long". Just getting hardware and software up-and-running can take up to 4 to 6 months. Two thirds of organization fall behind schedule when deploying new IT projects. And over half of those queried in the research experienced downtime for major infrastructure upgrades once projects were deployed. The result? You can’t deliver new business innovation at a pace that meets the demands of the marketplace.
Thus organization are creating new business-IT partnerships to apply technology in new, innovative ways; Rethink IT to help Reinvent Business
Main Point: While business and IT come to cloud for different reasons and with different goals, both roles are unified in their view of cloud’s overall value: the ability to deliver IT without boundaries, improve speed and dexterity, and create new business value. These are the kinds of value that bring participants together regardless of role. They speak to the marketplace drivers mentioned at the beginning of increasing expectations, demands, and competition. And they reflect the need to both rethink IT and reinvent the business. ------- Businesses are looking to see how they can get business advantage, and a cloud inspired approached to IT is a way of delivering that advantage. Social media delivered as a public cloud service Mobile working Agility to enable business change. Innovator or distruptor. Address historical challenges of it. Our world is changing dramatically whether we like it or not. Simply put, we’re being forced to rethink the way we run our business. First and foremost, Information Technology is becoming the heart of innovation and competitiveness. As a result, we cannot continue to add random complexity with isolated “solutions du jour”. New toys that neither reduce complexity nor have targeted value just won’t cut it! As core technology providers, we’re seeing the lines of business in every industry doing what they deem necessary to accelerate ability to compete - going to providers who promise fastest access to solutions. These providers are offering standardized infrastructures to reduce complexity, which we refer to loosely as Clouds. Delivered via Saas models, these are providing competitive alternatives to classic IT. That’s wonderful, except degrees of control around key elements such as securing infrastructure all of a sudden become that much more complex. Security becomes absolutely crucial to flexibility and innovation. At the same time, cost structures are forcibly collapsing, leaving us unable to address today’s pressure with business as usual approaches. We’re being forced to change. Regardless of what we call it, we must SIMPLIFY, STANDARDIZE and AUTOMATE our processes and tools. We must align ourselves with today’s technology transformations….OR… risk being sidestepped!
Flexibility in delivering , integrating and consuming cloud capabilities is enabled by open standards, delivered thur a common cloud platform. This allows clients flexibility and choice in how they consume cloud technologies With greater confidence that things will work together with a degree or portability….enabling a hybrid approach. ---- IT departments need to adapt to survive. Working with the business to adopting a new business model for IT, incorporating the best of what they already do with new delivery methods. That will include public cloud to deliver new application capabilities. Though is expected to be a blend of delivery approaches. This line illustrates IBM’s approach for the evolution of IT to remain relevant in todays world. You are here because you believe that there still is a role for in-house IT. IT must engage with exploitation of public cloud services, to ensure that the business is not exposing itself to unnecessary risk and providing guidance in adoption to avoid repeating failures of the past. IT can also evolve, evolving existing infrastructure to cloud delivery models, incorporating PureSystems to bring step changes in capability. What is key though is remaining open to industry change. CSCC. I’m not going to talk about open standards here, but there a chart in the deck to provide more detail. A key consideration in moving forward is how to avoid becoming locked into a new proprietry eco-system, say around a particular hypervisor. Competition in the hypervisor space is bringing change, need to be open to adopt it. Andy Jeknins discussed the virtualisation marketplace yesterday, Hyper-V is becoming mature, along with KVM becoming accepted for production workloads. KVM has been hindered by its lack of management tools, though this consideration is reducing with the wrapping of KVM with cloud mgmt tools. Many customers we are talking two are considering two hypervisors. Tooling has to support this trend. Also the tools that fit around the hypervisor and allow portability and choice of best of breed tools. Today, we are helping our clients with their cloud journey by operating their cloud environments or helping them build one behind their firewall. The IBM SmartCloud environment can have an order of magnitude impact on our clients operating costs all the while shifting them to an environment where they pay for only what they consume. In doing so, we introduce standardization and interoperability that empowers our customers and yields enormous productivity savings on an ongoing basis. And with that savings, we can help free up our clients to attack new channels, invest in social media and mobile, transform their front office, allow investment in bulletproofing their infrastructure, and help them reach new markets. All of these capabilities exist today in IBM. And with a deep understanding of our individual clients, we can help build a roadmap of where they can go and how to get there.
Benefit is more than just Though to DevOps. Let me share with you an example of one of our clients that are putting these capabilities to use. Citigroup. http://w3-01.ibm.com/sales/ssi/cgi-bin/ssialias?infotype=CR&subtype=NA&htmlfid=0CRDD-8F3QVU&appname=crmd Business Need: Citigroup wanted to dramatically reduce time to market by rapidly accelerating development cycles for the company ’s more than 20,000 internal application developers, who were typically forced to wait up to 45 days for server resources to be provisioned. Action Citigroup built an internal cloud using IBM® Cloudburst™ and Tivoli® software solutions, enabling self-service request, automated provisioning, and internal chargeback capabilities, while at the same time boosting utilization rates and improving operational efficiencies. Benefits: With the IBM solution, Citigroup slashed server provisioning times from 45 days to less than 20 minutes, speeding development cycles and allowing the company to put new features and enhancements in the hands of customers more rapidly. ---------------------------------- Long Form: Business need Citigroup Inc. has a division called Citi Technology Infrastructure (CTI), which serves as the company ’s IT backbone. CTI is responsible for improving customer experiences through thousands of new development initiatives each year. This division is in charge of more than 60,000 physical and virtual servers located in 14 different data centers. These servers include 8,500 development servers that are used by more than 20,000 internal application developers. As a result of working with a physical infrastructure, these developers have waited as long as 45 days to receive a requested provisioned server. In 2010, Citigroup began looking for a smarter approach to deliver these servers. The company wanted a solution that would automate and expedite the procedure for requesting and provisioning development servers for writing and testing code. Through this new solution, the company hoped to drastically reduce time to market by rapidly accelerating development cycles. Solution implementation After a successful proof of concept (POC) using the IBM WebSphere CloudBurst appliance (now IBM Workload Deployer appliance), Citigroup decided to implement a private cloud based on that architecture. Instead of taking a staged approach to hardware and software provisioning, the client ’s cloud deployment offers both infrastructure and software as a service (SaaS) from the outset. This architecture gives Citigroup the flexibility to optimize service delivery in numerous areas over the long term. The solution enables image-based standard operating environment (SOE) deployments with predefined middleware stacks, using VMware ESX software running as the hypervisor on a bank of Intel processor-based commodity servers. The Workload Deployer appliance allows Citigroup to create application environments simply and repeatedly for safe deployment and management within its private cloud. Plans are underway to extend cloud deployment to the existing IBM zEnterprise 196 server running the Linux on System z operating system. Once the deployment to the zEnterprise 196 server is complete, the client believes it will be able to offer the world’s fastest private cloud. To support service delivery processes, the team selected a suite of IBM Tivoli automation software solutions. IBM Tivoli Service Automation Manager software automates Citigroup administrative tasks, while IBM Tivoli Provisioning Manager software manages the provisioning workflow. IBM Tivoli Usage and Accounting Manager software enables usage-based chargeback to the relevant department, a process that encourages timely decommissioning of servers that Citigroup no longer requires. To help ensure security and continuous compliance in the cloud, Citigroup relies on IBM Tivoli Endpoint Manager software. Built on IBM BigFix technology, Tivoli Endpoint Manager software provides unified, real-time visibility and enforcement for software patches by installing an intelligent agent on each endpoint. Benefit of the solution By implementing IBM Tivoli automation software and an IBM Workload Deployer appliance, Citigroup became more efficient and slashed provisioning times from 45 days to less than 20 minutes. Requests for development servers are now handled through an internal self-service portal called Citi Marketplace. Here the client ’s thousands of employees can order everything from staplers to smart phones. These requests set off a secure, automated process for provisioning resources in the cloud. This process has dramatically improved systems administrator ratios, helping to reduce operational overhead and boost use rates. A systems administrator who supported 50 servers in the physical environment can now support more than 600 servers in the cloud. Further, the faster development cycles give Citigroup the ability to put new features and enhancements in customers’ hands more rapidly. The resulting transformation has been significant and popular. Citigroup has provisioned more than 550 virtual machines in the cloud so far, and demand for the new service continues to grow. Citigroup is taking full advantage of the cloud development environment to support a major development project for its retail banking experience. Client quote: “ We saw this as an evolution of IT infrastructure that goes from a dedicated world where servers are built to order to a virtual infrastructure in which the focus is on improving the use rates of our infrastructure. “The biggest operational overhead is systems administration time. Tools like IBM Tivoli Endpoint Manager automate functions that otherwise take up a considerable amount of time. “It lets our developers launch new features and enhancements more rapidly, which improves our customer experience in a shorter timeframe. You want developers coding as soon as possible, not waiting for servers. With our development cloud, when a new project comes along, we’re ready.” -- Graham Hill, Citigroup senior vice president “Time to market was one of the key complaints that we got from the development community. It forced us to look at our process of server provisioning, and this is when we decided to place our development in the cloud. With this implementation, we’ve reduced provisioning time from 45 days to less than 20 minutes, after approvals.” -- Jonathan Moore, Citigroup senior vice president “The doors have just been opened. Certainly people who are provisioning virtual machines or requesting virtual machines for development are moving to this as soon as they know it’s available. It’s just a North American initiative right now, but we’ve got people around the world knocking on the door.” – Jason Bisson, Citigroup vice president Solutions/Offerings Software: Tivoli: Tivoli Provisioning Manager; Tivoli Service Automation Manager; Tivoli Usage and Accounting Manager; Tivoli Endpoint Manager WebSphere: WebSphere CloudBurst App
Before embarking on a cloud journey, a productive first activity is to One of our key differentiators, although not a traditional managed service, are the analytical capabilities that we have within our strategy practice. This differentiated set of capabilities help us engage with you in a very productive conversation. It helps you, the client, understand your workloads, understand your IT and application environment. Its lets you understand the mix of simple applications, moderately complex applications and workloads and complex applications. What this does is it helps you to determine what targets can I move to what kinds of platforms . We perform an “as-is” assessment by conducting a brief question and answer session with you. With these inputs, we simplify an otherwise complex problem to filter out what projects will be of the most benefit to you and model that on the gain versus pain chart on the right. In this chart—which is based on one of our client engagements—the results show that development and test has a relatively low effort with a very high benefit (in the top right-hand corner) and so would be a primary choice to start with in terms of a cloud implementation. It also shows that there are other workloads, such as Internet applications, that have very high benefits for the organization but will also be more difficult to transfer (top left corner). With this workload analysis, developed by IBM Research, the client might decide there's some great work that can go and run on IBM SmartCloud Enterprise or there are some traditional heavily complex applications that can move to our managed applications services or to our managed service services because it's a highly complex IT infrastructure environment.. Note to presenter: Client example A financial client in North America was undertaking a companywide expense reduction program. It wanted to research the benefits of cloud computing. The business routinely creates development road maps for its organization; however, formulating a strategy and road map for six domains would typically take up to 72 weeks if done using existing methods. And the complexities of incorporating a cloud solution may have extended this planning even further. Over the next seven weeks, IBM worked with the client to create the road map. The client is now confident that the individual projects being executed will not only yield process improvements but also offer strategic value for the overall business.
Main point: Designed for the cloud, PureFlex systems are based on open architectures for ease of integration with existing environments. The result is a simplified experience, from system setup and solution deployment and through to day-to-day management and ongoing enhancements and upgrades. We have an image library of existing workload profiles that can be deployed. Secure access is ensured showing users only those workloads they are authorized to deploy. Workloads can be captured automatically and cataloged. PureFlex dynamically discovers the available compute, network, and storage resources available. PureFlex then analyzes the workload patterns and applies any provided policies about the user and workload. It then optimizes the available resources and deploys the workload rapidly with minimal effort.