Venyu Blog

There’s a part of me that believes, acutely, that learning mostly occurs in one direction, from the example (e.g. a mentor, a book, a professor), to the student. Again, that doesn’t take into advantage things like life experiences or intuition, however, more or less and for better or worse the diffusion and transfer of knowledge is mostly uni vs. bi-directional.

I believe, or at least until I previewed and read through this article, “Is Cloud Computing for Real?” which presented and summarized Jill Tummler Singer’s – CIO for the National Reconnaissance Office (NRO), at the recent Cloud Expo in NYC, that it was always the case, especially when it came to cloud computing. The private sector not only held all the cards, they also controlled the deck, their shuffle and how they were ultimately passed out to players, particularly in the public or government sector.  But there are some points Singer makes that I think can be applied as a construct to understanding (and overcoming) the challenges associated with any organization considering the cloud for virtual data storage.  Among these:

“As a CIO, building your cloud strategy can be confusing as there are many options to consider before making a decision. Cloud Computing can deliver within a variety of characteristics. Each CIO will need to consider the characteristics of primary importance in order to properly center a strategy. For example, determining your needs for rapid elasticity, on-demand self-service, global access, and payment options will be necessary to develop the right strategy. The second consideration for a CIO is just how far up the stack she wants to go. Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS) each deliver different features and capabilities for your business. You can develop a plan that includes some or all of the Services. Finally, a CIO needs to think about the deployment model(s) that best fit her business (Public, Private, Community, and Hybrid).”

And this:

“Our cloud delivery model will be private first and then a hybrid of private and community (with our Intelligence Community colleagues). We’re moving our way up the stack starting with IaaS and PaaS first, then SaaS. And, we are focused on a phased revolutionary approach—building a new environment as a green field…Defining specific roles and responsibilities, articulating cloud service level agreements between the provider and users, adjusting organizational constructs to fit the Cloud business model, and streamlining configuration control processes to meet the rapid timelines of provisioning must all be addressed if success is to be achieved.”

I found the keystone to her entire presentation, however, was vested in and well-served by the following slide which illustrates how the NRO “Cloud” will be implemented over time.

As Singer describes them this “phased approach” includes requirements paced to meet federal budget processes, existing acquisition commitments, and mission service imperatives including:

Phase 1 – Test It. We have multiple small cloud pilots ongoing covering commodity/utility cloud, high performance cloud, and big data cloud. We are focused on streamlining information assurance compliance, developing appropriate policies and processes, and defining standards.

Phase 2 – Prove It. In this phase, we will merge successful elements from Phase 1, finalize policies, prototype management and governance changes necessary for cloud provisioning models, refine standards, realign programs as needed, and ensure enterprise foundation capabilities meet mission demands.

Phase 3 – Use It. Scaled deployment of the enterprise Cloud will occur. Applications will develop specific transition and migration plans generally aligned with planned refresh cycles. Service management processes and new centralized information assurance approaches will be refined for long-term success.

Phase 4 – Exploit It. The NRO Cloud will emerge as the primary Enterprise Infrastructure Services Provider with robust measures of effectiveness, performance accountability, and service transparency. New mission and business applications will emerge through innovations in Cloud Computing and federation with the IC Cloud will be achieved moving the NRO into a hybrid environment.

In fact, take out the references to the NRO Cloud and these phases, cumulatively, could easily be applied to all of us here in the DPS (e.g. dreaded private sector). Additionally, Singer’s conclusions articulate the true duality of the “cloud” as a valuable tool for privately-held as well as publicly managed organizations including the federal government.

In summary, the time for Cloud Computing is now. All of the indicators are in place to suggest a rapid and successful migration to this newest generation of IT architecture. It is not a one-size-fits-all migration. CIOs will need to spend time exploring Cloud Computing solutions to find the combination that best responds to specific mission and business needs. Maintaining awareness of the concerns and mistakes is absolutely necessary and security remains the top concern. Security is a huge focus for IT vendors and creative, robust solutions will be constantly emerging to reduce concerns. Large organizations will increasingly turn to Enterprise Cloud Computing strategies to achieve efficiencies of Cloud Computing while reducing risks of exposure for intellectual property, customer privacy, and competitive strategies.

I guess it is true that in really rare cases, learning can indeed flow back from the student to the master.