The Private Cloud Journey

“Private cloud adoption is a journey both from a technical and business perspective.” At the recent AFCEA Cloud Lifecycle Management Symposium in DC, the discussion on government cloud computing ranged from acquisition policies to building the roadmaps in which NIST and government guidelines are being centered around. The vision of these roadmaps is to “easily locate desired IT services, rapidly procure access to the services, and use the services to deliver innovative mission solutions.” But with all of the service providers and offerings available, how can government standardize and corral all of these into one simple menu of options that meets individual agency requirements? How will agencies define a successful cloud program? What are the strategies to assure success? This event geared towards identifying systems and solutions for these for cloud quandaries as agencies move closer to the deadlines set forth in the “Cloud First” policy. The value of cloud has started to become something tangible now but implementing is still a major roadblock. Government has reached the proverbial fork in the road between where real mission-critical needs meet technical resource demands. The following is an excerpt on pre-requisites, prepared by David Blankenhorn, head of the DLT Cloud Advisory Group, for CIOs to consider when enabling a future private cloud platform. DLT recently launched this cloud knowledge practice to provide advice and guidance to government customers about all of their options to develop a clear and actionable plan for which resources they need to acquire. The silver lining to all of the conversation is that it is indeed possible to begin to realize some of the benefits of a cloud platform by taking incremental steps:
  • Standardization:   Standardization is the key to reducing complexity, and by extension, administrative costs.  It may be more appropriate to define the standards for all new services moving forward, rather than trying to retro-fit legacy systems.  Key areas of focus should be on Standardized Operating Environments (SOE) for the operating platform and Service Oriented Architecture (SOA) for the applications.
  • Skill Shifts: Managing a cloud environment will require new skills or enhancements to existing skills.  A deep understanding of virtualization (network storage, and server), performance analysis, capacity planning, and automation (provisioning, orchestration, and metering) will all be needed.  Additionally, key IT disciplines like Service Level Management and being able to understand, develop and negotiate Service Level  Agreements (SLA) will be critical.
  • Virtualization: While virtualization does not make a cloud, virtualization is a key enabler for building cloud platforms.  If not already in place, consider virtualizing some existing systems and applications.
The Seed Cloud: Another possible approach is to develop a “seed cloud.”  Begin by looking for new projects that have common architectures like databases or application servers.  Rather than purchasing assets specific to each project, consider pooling the IT asset budget for these projects to build out a seed cloud that will support all of these projects.  This seed cloud can serve as a foundation for growing out the private cloud service offerings.